Add Chapter 3 and 6 folders
Browse filesAdds the full Chapter 3 and Chapter 6 folder contents from the GitHub notebooks repository.
The folders include notebooks, JSON loss files, generated plots, and supporting figure source files.
- Chapter 3/balanced_knapsack.png +3 -0
- Chapter 3/balanced_knapsack_packing.png +3 -0
- Chapter 3/chapter_3_batched_loss.json +1 -0
- Chapter 3/chapter_3_loss.json +1 -0
- Chapter 3/chapter_3_loss_vizualisation.ipynb +0 -0
- Chapter 3/chapter_3_padding_vizualisations.ipynb +0 -0
- Chapter 3/constraint_padding.png +3 -0
- Chapter 3/create_fig_3_1.py +220 -0
- Chapter 3/create_fig_3_2.tsx +221 -0
- Chapter 3/greedy_knapsack_packing.png +3 -0
- Chapter 3/knapsack_packing.png +3 -0
- Chapter 3/loss_comparison.png +3 -0
- Chapter 3/loss_plot_first_5000_steps.png +3 -0
- Chapter 3/minimal_vlm_training.ipynb +2307 -0
- Chapter 3/naive_packing.png +3 -0
- Chapter 3/naive_padding.png +3 -0
- Chapter 6/chapter_6_loss_vizualisation.ipynb +0 -0
- Chapter 6/cross_attention_batched_loss.json +1 -0
- Chapter 6/cross_attention_batched_loss_256_image_tokens.json +1 -0
- Chapter 6/loss_comparison_attn_vlms.png +3 -0
- Chapter 6/loss_comparison_attn_vlms_same_tokens.png +3 -0
- Chapter 6/self_attention_batched_loss.json +1 -0
Chapter 3/balanced_knapsack.png
ADDED
|
Git LFS Details
|
Chapter 3/balanced_knapsack_packing.png
ADDED
|
Git LFS Details
|
Chapter 3/chapter_3_batched_loss.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
[2.2656, 1.6328, 1.6016, 2.1406, 1.2578, 1.4375, 1.9141, 1.0234, 1.4922, 1.2969, 1.3906, 1.7812, 1.0703, 1.9297, 1.3516, 1.6719, 1.3203, 1.7266, 1.3359, 1.3516, 1.1328, 1.7109, 1.5234, 1.375, 2.2656, 1.625, 1.2422, 1.4375, 0.8125, 1.5859, 1.3672, 1.1953, 1.1016, 1.3984, 1.3047, 1.1797, 1.9844, 1.5781, 1.1875, 1.375, 1.7969, 0.7852, 1.4766, 1.3984, 1.4609, 1.2734, 1.7266, 1.3984, 2.3125, 1.3906, 1.3516, 1.5312, 1.0625, 1.3047, 1.7969, 1.6484, 1.1953, 1.4766, 1.7266, 1.6094, 1.7578, 1.7031, 1.2656, 1.9219, 1.1406, 1.1875, 1.4766, 1.5625, 1.5312, 1.3203, 0.9688, 1.6328, 1.0625, 1.25, 0.9375, 1.8672, 1.3984, 1.9609, 1.4531, 1.1641, 1.7578, 1.6016, 1.6953, 1.7031, 1.3203, 1.4531, 1.8125, 1.9062, 1.9453, 1.5625, 1.4141, 1.125, 1.1484, 1.2812, 1.6172, 1.7266, 1.1562, 1.7344, 1.3594, 1.0312, 1.2891, 1.3672, 1.125, 1.0547, 1.5469, 1.5312, 1.2422, 1.3438, 1.2656, 1.5625, 1.7812, 1.3203, 1.6172, 1.6875, 1.2422, 1.4844, 2.3125, 1.7734, 1.625, 1.1016, 1.1484, 1.2109, 1.3594, 1.6172, 1.6562, 1.3594, 1.0078, 1.4688, 1.0625, 1.6328, 1.2109, 1.4844, 1.1172, 1.125, 1.1328, 1.3906, 1.5156, 1.7188, 1.4688, 1.3672, 1.1797, 1.6406, 1.375, 1.2812, 1.875, 1.4375, 1.4297, 1.3203, 1.1641, 1.4609, 1.0, 1.0781, 1.7422, 1.4141, 0.9531, 1.5234, 2.125, 1.4922, 1.6328, 1.875, 1.1875, 1.1406, 0.9766, 1.7266, 1.1953, 1.2656, 1.0781, 1.6953, 1.6328, 1.5156, 1.4141, 1.9453, 0.957, 1.1797, 1.3516, 1.4844, 1.4766, 1.125, 1.9453, 1.2188, 1.1562, 1.3125, 1.5, 1.3125, 1.4219, 1.6016, 0.9883, 1.0469, 1.3047, 1.6172, 1.0625, 1.6016, 1.5938, 1.4375, 1.3672, 1.1094, 1.2656, 1.0234, 1.375, 1.6406, 1.1016, 1.2578, 1.3984, 1.1016, 1.8203, 1.5547, 1.3672, 1.2266, 1.0156, 1.3203, 1.3125, 1.5781, 1.6797, 1.4297, 1.1797, 1.4219, 1.2422, 2.4375, 1.2109, 1.625, 1.4219, 1.4688, 1.2109, 1.0312, 1.5469, 1.2578, 1.3203, 1.7891, 1.4062, 1.1641, 1.2109, 1.3672, 1.9141, 1.5391, 1.8281, 1.6562, 1.2266, 1.7344, 1.0391, 1.3672, 1.1328, 0.9922, 1.3906, 1.2031, 1.2344, 1.0156, 1.1719, 1.3672, 1.9609, 1.6641, 1.2266, 1.2891, 1.0781, 1.6406, 1.7891, 1.2109, 0.8867, 1.5859, 1.3828, 1.4141, 1.4766, 1.0156, 1.1328, 0.9297, 1.2188, 1.5781, 1.4688, 0.9336, 1.1172, 1.3281, 1.1016, 1.3203, 1.375, 0.8906, 1.5625, 1.6719, 1.6328, 1.1328, 1.4453, 1.4688, 0.8516, 0.8125, 1.2031, 1.4453, 1.0703, 1.1719, 1.1172, 1.3516, 1.3125, 1.3359, 1.0469, 1.0703, 0.9805, 1.8516, 1.0547, 1.6094, 1.0469, 1.25, 1.5234, 1.9219, 1.3047, 1.9531, 0.9609, 1.1406, 1.3906, 0.9805, 0.793, 1.5078, 1.5625, 1.3594, 1.6016, 1.2031, 1.3516, 1.4062, 1.2188, 1.5938, 0.9492, 1.4922, 1.6328, 1.2812, 1.3438, 1.0078, 1.625, 0.9297, 1.0859, 1.2578, 1.3203, 1.3438, 1.3047, 1.2891, 1.5391, 1.2578, 1.3281, 1.5859, 1.5625, 0.9727, 1.4297, 1.5859, 1.8516, 1.4844, 0.9102, 1.4141, 1.2969, 1.3203, 1.2344, 1.8906, 1.9609, 0.8867, 1.6328, 1.3281, 1.4844, 1.1562, 1.1953, 1.1641, 1.0469, 0.9219, 1.7266, 0.9766, 1.4766, 0.7578, 2.0781, 1.2656, 1.2422, 1.3906, 1.8594, 0.7734, 1.2969, 1.125, 1.2031, 1.3828, 1.3516, 1.0391, 1.0, 0.9609, 1.5938, 1.5312, 1.1953, 1.1406, 1.9219, 1.1016, 1.3984, 0.9531, 1.5156, 1.2422, 1.2656, 1.1562, 1.0469, 1.1094, 1.0859, 1.0703, 1.3203, 1.4375, 1.6172, 0.832, 1.2422, 1.3359, 1.3906, 1.0391, 1.4531, 1.0938, 1.6328, 1.75, 1.6875, 1.4922, 1.3359, 0.7969, 1.1484, 1.4688, 1.0234, 1.1406, 0.9219, 1.6484, 1.3594, 1.5469, 0.957, 1.8438, 1.2344, 0.8008, 1.4375, 1.2578, 0.7031, 1.4688, 1.3359, 1.2812, 1.2109, 1.3906, 1.0, 1.0781, 1.2109, 1.2578, 1.0781, 1.1875, 0.582, 1.2188, 1.2031, 1.3906, 1.0078, 1.3984, 1.1797, 1.5312, 1.1641, 1.7891, 1.3281, 1.7109, 1.0312, 2.4219, 1.5625, 1.0, 1.2188, 2.0469, 1.8203, 1.1641, 1.1016, 1.1719, 0.8086, 1.7734, 0.7891, 1.2188, 1.2031, 1.3594, 1.3906, 1.2344, 1.0547, 2.0469, 1.1562, 1.6953, 1.1953, 1.7109, 0.9727, 1.1641, 1.2188, 1.0156, 0.7188, 1.3828, 1.7188, 0.8945, 1.5625, 1.7031, 1.25, 1.4141, 1.1641, 0.8125, 1.7891, 1.2734, 1.0781, 1.125, 0.6914, 0.9961, 1.0703, 0.8438, 0.918, 0.8477, 2.0312, 1.1797, 0.8867, 2.2812, 0.9766, 0.7383, 1.0234, 1.2891, 1.6094, 1.4766, 1.5547, 1.1562, 1.6172, 1.0156, 1.4844, 0.918, 1.1328, 1.4688, 0.8711, 1.4766, 0.8555, 1.6484, 1.6641, 1.2188, 1.0859, 1.4375, 1.3125, 1.2422, 1.5312, 1.3047, 1.3047, 1.5625, 1.0312, 1.6094, 1.1875, 1.6328, 1.1328, 1.3984, 1.1719, 1.0781, 0.625, 1.3359, 1.25, 1.4062, 1.4922, 0.9648, 0.7383, 1.2266, 1.4062, 1.3125, 1.375, 1.0781, 1.5938, 0.6797, 1.4922, 0.8008, 1.4219, 0.918, 0.7891, 1.625, 1.2734, 1.4844, 1.1797, 1.9062, 0.7734, 1.7031, 1.1094, 1.0703, 1.3203, 1.6328, 0.9062, 1.1875, 1.2969, 1.8906, 0.8867, 1.4219, 1.2031, 1.4453, 1.4531, 1.3594, 1.7812, 1.4219, 1.0781, 1.1484, 1.5391, 1.4844, 1.5547, 2.0156, 1.4297, 0.8828, 1.125, 1.2578, 1.4375, 1.4062, 1.4609, 0.9531, 1.25, 1.2812, 1.5391, 1.2344, 1.5234, 1.4062, 1.1875, 1.3594, 1.2344, 1.2344, 1.2812, 1.0781, 1.3359, 1.4844, 1.5234, 1.1016, 1.0781, 1.1562, 0.7734, 1.4297, 1.4609, 1.5781, 1.2578, 1.0391, 2.0625, 1.3359, 1.3359, 1.7969, 1.4062, 0.8828, 1.4062, 1.3906, 1.1641, 1.5312, 1.2891, 1.0156, 1.4141, 1.1172, 1.3047, 1.1172, 1.2969, 1.3203, 0.9922, 1.8359, 1.4844, 1.2266, 1.2812, 1.5078, 1.25, 1.3281, 0.7773, 1.0, 1.2422, 1.3984, 1.5078, 1.1016, 1.1562, 1.3672, 0.875, 1.7891, 0.7148, 1.4297, 1.7422, 1.7031, 0.8164, 1.2422, 1.3203, 0.8633, 1.4531, 1.2578, 1.2891, 1.0547, 2.125, 1.2578, 1.7344, 1.4141, 1.0781, 1.0469, 1.0938, 1.1484, 1.3906, 1.1719, 1.4922, 0.8359, 1.0938, 1.0312, 1.1172, 1.3828, 1.1875, 1.6641, 0.9766, 1.5938, 1.1406, 1.3672, 1.0703, 1.3203, 1.1016, 1.5078, 0.7227, 1.7344, 0.8984, 1.4141, 1.3516, 1.5625, 0.6484, 1.5078, 1.0547, 1.4531, 1.2344, 1.1172, 0.6797, 1.4375, 1.2734, 1.6953, 2.1094, 1.1953, 1.2656, 1.2188, 1.3828, 1.2656, 1.6953, 0.8438, 1.4375, 1.3672, 1.5469, 1.25, 1.0, 1.2656, 1.4297, 1.4922, 1.4297, 1.5625, 1.2188, 0.9844, 1.0078, 1.7812, 1.7422, 1.2344, 1.4609, 1.625, 1.1172, 0.7773, 1.4531, 0.8672, 1.2109, 0.9453, 1.2891, 1.4375, 1.8203, 0.9102, 1.5234, 1.8281, 1.2031, 1.1562, 1.1953, 0.8594, 0.9648, 1.2969, 1.1562, 1.4375, 1.4766, 1.5547, 1.1406, 1.2031, 1.5469, 1.6641, 1.4141, 0.8359, 1.4453, 1.25, 1.6641, 1.3516, 0.8789, 0.9023, 1.5703, 1.2734, 1.4453, 1.3125, 1.1328, 1.3047, 0.8203, 1.4688, 0.9414, 1.2891, 1.2109, 0.8242, 1.1328, 1.25, 1.9375, 0.7852, 1.6016, 0.9102, 1.2812, 1.0234, 1.6094, 1.0469, 1.4609, 0.8672, 0.9297, 0.8125, 1.3125, 1.125, 1.8359, 1.3203, 1.2188, 1.6016, 0.668, 1.1875, 1.5547, 1.5938, 1.2266, 1.4062, 1.3125, 0.8711, 0.8164, 1.5469, 1.1719, 1.1641, 1.1484, 1.2266, 1.1641, 1.5781, 1.5234, 1.4922, 1.1562, 0.9102, 1.7422, 1.8203, 0.7969, 0.9922, 1.5703, 1.5156, 1.2578, 1.2422, 1.2578, 0.9805, 1.0391, 1.3984, 1.1719, 1.2031, 0.7188, 1.7344, 0.6445, 1.4531, 0.7031, 1.0703, 1.6719, 1.4766, 1.2656, 1.3906, 1.125, 1.2344, 0.75, 1.7891, 0.957, 1.2031, 1.7266, 1.9531, 1.2891, 1.7188, 1.4219, 1.6484, 0.9336, 1.4062, 1.2109, 1.1797, 0.8359, 1.0, 1.5, 1.0938, 1.2344, 1.4375, 1.4141, 1.3516, 1.6484, 1.1641, 0.7891, 0.7852, 1.3203, 0.9883, 1.25, 1.625, 1.0, 1.3906, 1.0312, 1.3516, 1.2969, 1.4062, 1.4609, 1.5859, 1.4219, 1.0859, 1.2891, 1.3984, 1.3359, 1.1328, 0.8711, 0.8516, 1.5859, 1.5234, 0.8398, 0.793, 0.9492, 1.6641, 1.1094, 1.1172, 1.3828, 1.2578, 1.3516, 1.5625, 1.1094, 1.3672, 1.7812, 0.9648, 1.2656, 1.6953, 0.9141, 1.5938, 0.9414, 1.1094, 1.0938, 0.6953, 0.9062, 0.8672, 1.5625, 1.1562, 0.8086, 0.7461, 1.8359, 0.7539, 1.7344, 1.6016, 1.2109, 1.125, 1.0781, 1.4141, 0.957, 1.8047, 1.5547, 1.0703, 1.6328, 0.9375, 1.0625, 1.2891, 0.9297, 0.9844, 0.8125, 1.4297, 1.2969, 1.2344, 1.2031, 1.5469, 1.2812, 1.6328, 0.8164, 1.1875, 1.1562, 1.1172, 1.4531, 1.1562, 1.2109, 0.7422, 1.6719, 0.9883, 1.5469, 1.2109, 0.7891, 0.7305, 1.3672, 0.9805, 1.3906, 1.2344, 1.6406, 1.6094, 1.0469, 1.3359, 0.9414, 1.3203, 1.2344, 1.1484, 1.3047, 0.8359, 1.2031, 1.375, 2.0156, 1.5781, 1.7812, 1.5859, 1.2812, 0.7148, 1.0781, 1.2188, 0.9414, 0.8164, 0.9414, 1.5938, 1.2656, 0.957, 1.0938, 1.1094, 1.2734, 1.5078, 1.6406, 1.1953, 0.9805, 1.7344, 0.9688, 1.3594, 1.2188, 1.2266, 1.4609, 1.4062, 1.1328, 1.6328, 1.5391, 1.2031, 0.8516, 1.0781, 1.0078, 0.7109, 1.8438, 0.8711, 0.7578, 1.0234, 0.9805, 1.6094, 1.1094, 0.9336, 1.2109, 1.7422, 1.2969, 0.7734, 1.6094, 1.3125, 1.2031, 1.0625, 1.0781, 1.5234, 1.1953, 1.3047, 0.9805, 0.75, 1.1484, 1.4609, 1.0469, 0.8594, 1.1484, 1.6328, 1.0703, 1.0625, 1.1953, 1.3359, 0.7773, 1.7266, 1.2344, 0.9297, 0.9062, 1.3125, 1.3359, 0.9414, 0.7344, 1.0391, 1.4531, 1.7578, 1.4062, 1.5547, 1.0156, 1.7422, 1.1328, 1.3672, 1.0547, 1.4062, 0.8477, 1.4453, 1.5156, 1.4688, 0.9492, 1.2734, 1.2031, 1.1719, 1.0703, 0.8828, 1.0312, 1.2891, 1.5391, 0.9766, 0.9297, 1.4922, 1.2109, 0.8867, 1.4453, 1.1406, 2.1406, 0.6953, 1.375, 0.8359, 2.0938, 1.3906, 0.9414, 1.4609, 1.1094, 1.1172, 1.2422, 1.3672, 0.7344, 1.1562, 0.8359, 0.8789, 0.9648, 0.8867, 0.8398, 1.1484, 0.9766, 1.4609, 1.0391, 1.0625, 1.3906, 1.7188, 1.2891, 0.8438, 1.125, 1.6719, 1.1797, 1.9531, 0.8594, 0.7734, 1.7031, 1.1641, 1.5703, 1.6406, 1.7109, 1.3438, 0.7969, 1.2188, 1.9141, 0.6914, 1.5469, 1.0, 1.2422, 1.0547, 1.4297, 1.5312, 1.8594, 1.2891, 0.9688, 0.9922, 0.9609, 0.8867, 0.8438, 1.4375, 1.2812, 1.2578, 1.3359, 1.0703, 0.9609, 1.1172, 1.5469, 1.2109, 1.0625, 1.4141, 1.0156, 1.3359, 0.9609, 0.9023, 1.4141, 1.0469, 0.8359, 1.6484, 1.3281, 1.2656, 0.8516, 1.7266, 1.2656, 1.1562, 1.1875, 1.8516, 0.9219, 1.3516, 1.1719, 1.0859, 1.3125, 1.4688, 0.8789, 0.9609, 1.0391, 1.6797, 1.2031, 1.6172, 0.8789, 1.0703, 1.2969, 1.2734, 1.1406, 1.5781, 1.3984, 1.1953, 0.9219, 1.4844, 1.7188, 1.0234, 1.3906, 1.2109, 0.8672, 0.7969, 1.0547, 1.0547, 1.2891, 0.9805, 0.9453, 1.1641, 1.3828, 1.1406, 0.8281, 1.3594, 1.0703, 1.1406, 1.9219, 1.5547, 1.5859, 0.9883, 0.8164, 1.1328, 1.5547, 1.2812, 0.9648, 1.8672, 1.0, 1.0234, 1.0703, 1.125, 1.1328, 1.3438, 1.3828, 1.1328, 1.0078, 1.1406, 1.7734, 1.1172, 0.8242, 0.957, 0.9531, 1.8516, 0.7383, 0.6875, 1.1875, 1.6172, 0.8672, 0.7734, 1.0078, 1.2109, 0.9258, 1.2109, 1.3906, 1.1953, 2.2188, 1.4531, 1.4062, 1.3828, 1.6484, 1.3906, 0.8906, 1.1641, 1.1016, 1.1328, 0.8125, 1.1016, 1.2188, 1.0156, 1.2656, 1.5, 1.0234, 1.125, 1.0703, 1.3047, 1.8984, 1.1016, 1.1562, 1.2578, 1.5234, 1.4609, 1.4531, 1.3516, 0.9492, 0.875, 1.1875, 1.3516, 1.0859, 1.2188, 1.5312, 1.5312, 0.8398, 0.8477, 0.8633, 1.3047, 1.3438, 1.125, 1.3359, 0.8203, 1.2734, 1.8438, 1.5234, 1.6953, 1.0703, 1.0781, 0.9297, 0.8086, 1.1719, 0.8555, 1.6875, 1.0, 0.7852, 1.5938, 1.2266, 1.1953, 0.9297, 1.2969, 1.4766, 1.3672, 1.1875, 1.2031, 1.4766, 0.9492, 1.4609, 1.5703, 1.1406, 0.832, 0.9414, 0.6367, 0.918, 1.8828, 0.9961, 1.5547, 1.5391, 1.0391, 1.0625, 1.3516, 0.8828, 1.1562, 1.2969, 1.1484, 1.0, 0.6758, 0.7852, 1.7969, 1.7109, 1.5391, 1.3906, 1.1641, 1.7734, 1.2812, 1.4141, 1.3828, 1.5, 1.0469, 1.3047, 1.6094, 0.8789, 1.5391, 1.5625, 1.5234, 0.8867, 1.1328, 0.9609, 1.1641, 1.4844, 0.75, 1.4297, 0.9727, 0.9922, 1.0547, 1.3906, 1.2266, 1.3203, 1.375, 1.1484, 1.4141, 1.3672, 1.4688, 1.5312, 1.3828, 0.7266, 1.3047, 1.3125, 1.2109, 1.4766, 1.2031, 1.75, 1.0703, 0.9297, 1.4766, 1.6094, 0.8516, 1.4375, 1.0156, 0.8242, 1.5234, 1.1172, 0.6562, 1.5391, 1.0, 1.3828, 1.4141, 1.3438, 1.4297, 1.7031, 1.3516, 0.9883, 1.5078, 1.1094, 1.0938, 1.6797, 1.6328, 1.3672, 1.4375, 0.9531, 1.1953, 1.1328, 1.7422, 1.3047, 1.8594, 1.1406, 1.0078, 1.2578, 1.25, 1.0859, 1.3438, 1.0938, 0.6758, 0.9805, 1.1719, 1.125, 1.7656, 1.3906, 0.9688, 1.3672, 1.3672, 1.4453, 1.1641, 1.2656, 1.0781, 0.5742, 1.1875, 1.25, 1.0547, 1.5234, 1.1406, 1.0703, 0.875, 1.3281, 0.9023, 1.5859, 0.8008, 1.0625, 1.4453, 1.1328, 1.3984, 0.9922, 0.9453, 0.7852, 1.3281, 1.2969, 1.2812, 1.3594, 1.3672, 1.1953, 1.5703, 1.1406, 1.7266, 1.1875, 1.2266, 1.2578, 1.0, 1.5078, 1.3906, 2.2656, 1.3516, 0.7539, 1.2969, 1.1719, 1.1719, 0.8242, 1.4531, 1.3125, 1.3438, 0.9102, 0.8047, 1.0703, 1.6641, 1.3594, 2.2344, 1.4141, 1.3672, 0.8438, 1.0781, 1.1484, 1.0938, 1.5625, 1.6016, 1.1953, 1.0703, 0.9883, 1.3906, 1.0391, 0.6602, 0.6562, 1.4531, 1.3516, 1.4375, 1.2656, 1.3594, 0.8398, 0.9297, 1.4219, 1.4844, 1.4141, 0.8984, 1.2266, 0.9688, 1.2891, 1.0469, 1.2891, 1.125, 1.0312, 1.7422, 0.9336, 1.2578, 1.3672, 1.3125, 1.3438, 1.6484, 1.0859, 1.4922, 1.4531, 1.2188, 1.1328, 1.1875, 1.1641, 1.4922, 1.4141, 1.4844, 1.2969, 1.5469, 1.0234, 1.5156, 0.7656, 1.6797, 1.1562, 1.2812, 1.3672, 1.1953, 1.2578, 1.4297, 1.7188, 1.3438, 0.9805, 0.9727, 1.25, 1.3203, 0.9062, 1.3516, 0.9609, 0.8438, 1.3281, 0.7031, 1.0, 1.4141, 1.4844, 0.9219, 1.4922, 1.6172, 0.918, 1.1953, 1.3984, 0.8984, 0.9766, 0.793, 1.3047, 1.3281, 1.2266, 1.4141, 0.8438, 1.2109, 0.9102, 1.5234, 1.375, 0.7148, 1.1172, 0.9609, 1.6953, 1.1484, 1.2266, 1.4453, 0.8672, 1.2812, 1.25, 1.3203, 1.4141, 1.6172, 1.1641, 0.9062, 0.8828, 0.7812, 1.2188, 0.9844, 1.6484, 1.0781, 0.918, 1.7812, 0.957, 0.9414, 1.1406, 1.4609, 0.6562, 1.4141, 1.4219, 1.5078, 1.0391, 1.1719, 0.7383, 1.6484, 1.3906, 2.0469, 0.9844, 1.3281, 0.6953, 1.1406, 1.2031, 1.3203, 1.2031, 0.6562, 0.9258, 1.1719, 0.7383, 0.9375, 1.3672, 1.1016, 1.1484, 1.9766, 1.0234, 1.5703, 1.6406, 1.2891, 1.1797, 1.0234, 0.7109, 1.1172, 0.9805, 1.1719, 1.7422, 1.1797, 1.5, 0.9531, 1.25, 1.0859, 1.1016, 1.1406, 1.2734, 1.1562, 1.0938, 1.6328, 0.8203, 1.3672, 0.8867, 1.4141, 1.0781, 1.2422, 0.9609, 0.9102, 1.2734, 1.125, 0.6641, 1.1641, 0.7422, 1.4609, 0.875, 1.0703, 1.1797, 1.5938, 0.9883, 0.9297, 1.5859, 1.3516, 1.1641, 1.1797, 1.0312, 1.1719, 0.75, 1.5625, 0.7422, 1.0312, 1.3828, 1.5703, 1.3516, 1.4375, 1.4141, 0.6562, 1.4531, 1.3438, 0.7969, 1.4062, 1.2656, 1.9219, 1.6172, 1.0859, 1.6562, 1.4375, 0.9531, 1.3047, 1.4531, 1.5234, 1.3281, 0.8555, 1.2891, 1.1406, 1.4453, 1.6484, 1.4766, 1.5781, 1.3594, 0.9023, 0.6406, 1.2344, 0.8164, 0.8477, 0.9102, 1.7344, 0.8867, 1.3203, 1.1406, 1.125, 1.0469, 1.3516, 0.8359, 1.1328, 1.2266, 1.4766, 1.3828, 0.7109, 1.2422, 1.6094, 1.1641, 1.1719, 1.2969, 1.0391, 1.125, 1.5078, 0.7656, 0.9023, 1.5156, 1.1172, 0.832, 1.5859, 1.3516, 1.2812, 0.9141, 1.3438, 1.7656, 1.0859, 0.9414, 0.8867, 0.9062, 0.8945, 1.2422, 0.9258, 1.0078, 1.0078, 1.6953, 1.6094, 1.3125, 2.125, 0.8398, 1.3047, 1.3438, 1.2031, 0.9297, 0.9492, 1.125, 1.8359, 0.625, 1.5078, 1.3828, 1.1172, 1.2188, 1.1641, 1.3125, 1.25, 1.4375, 1.2656, 1.2344, 1.375, 0.7891, 0.9688, 0.8281, 1.2266, 1.0938, 1.1719, 1.0938, 1.3672, 1.4844, 1.1562, 1.7891, 1.2891, 1.375, 1.4062, 1.4922, 0.793, 1.1094, 1.3438, 0.7422, 0.918, 1.2734, 0.9375, 1.2734, 0.6328, 0.9336, 1.5859, 0.6992, 1.2109, 1.1484, 1.2891, 1.0, 1.9062, 1.0234, 1.5938, 1.0234, 1.0859, 1.1562, 0.8906, 1.0469, 1.2266, 1.375, 1.1719, 1.0469, 1.5391, 1.4531, 0.9219, 1.2109, 1.2109, 1.2891, 1.2734, 0.7031, 0.7148, 1.2891, 1.3906, 0.8086, 1.4141, 0.9492, 1.1641, 1.2969, 1.2422, 0.9766, 1.3516, 1.0859, 1.0, 1.4453, 1.3359, 1.0391, 1.0078, 0.6836, 0.7617, 1.25, 0.8086, 1.3438, 1.6641, 1.0547, 1.1172, 0.957, 1.3203, 0.8242, 0.9609, 0.7539, 1.1719, 1.3594, 1.6406, 1.1016, 1.3594, 0.7734, 1.2188, 1.1641, 1.0938, 1.5859, 1.1875, 1.1172, 1.3359, 1.1094, 1.0156, 1.3125, 1.4609, 1.1875, 1.0234, 1.2812, 1.5, 1.5469, 1.4844, 0.8555, 1.1875, 1.1797, 1.4766, 1.3672, 1.0625, 1.1719, 1.0312, 1.4766, 2.4062, 0.9688, 1.0859, 1.0625, 0.8398, 1.1719, 1.1562, 1.3203, 1.4531, 1.0625, 1.0703, 1.4375, 1.3828, 0.9414, 1.2188, 1.4062, 1.2188, 0.9375, 1.1953, 1.0703, 1.7734, 1.0078, 0.8281, 1.125, 1.4453, 1.125, 1.5, 1.3984, 1.1484, 1.375, 1.125, 0.6641, 1.0938, 1.3047, 1.2734, 0.918, 1.2109, 1.1719, 0.957, 1.4219, 1.2422, 1.2578, 0.7461, 1.1719, 1.375, 0.8359, 0.9531, 1.0625, 1.6719, 1.6016, 1.4297, 1.4219, 1.375, 1.1328, 1.5547, 1.2969, 1.5938, 1.3125, 1.7109, 1.25, 1.6094, 1.0625, 0.8711, 1.0078, 1.3906, 0.957, 1.0078, 1.4375, 0.8867, 1.25, 0.7422, 1.2578, 1.6797, 1.1719, 1.0938, 1.6797, 1.1016, 0.8242, 1.4453, 1.2969, 1.2812, 1.0469, 1.1172, 1.0469, 1.5312, 1.0156, 1.1641, 1.2422, 0.8398, 1.4609, 0.7227, 1.1484, 1.3047, 0.9102, 1.7109, 1.1484, 1.5859, 1.2969, 1.3047, 1.6328, 0.8477, 1.2422, 1.5469, 1.1328, 1.2578, 0.9336, 1.1719, 1.1562, 1.2969, 1.6953, 0.8633, 1.0547, 1.1172, 1.2188, 1.1953, 1.4531, 1.0859, 0.8516, 1.2734, 1.6562, 2.0938, 1.3438, 0.8672, 1.1484, 1.0469, 0.8555, 1.1406, 1.4219, 1.7734, 0.6133, 1.3672, 1.375, 1.2031, 1.2109, 1.4688, 1.3438, 0.6133, 1.2266, 1.3906, 1.2031, 1.3516, 1.1562, 0.8633, 1.1875, 1.0859, 0.9688, 1.1953, 1.2266, 2.0781, 1.2266, 1.3984, 1.0078, 0.9883, 0.8789, 1.3125, 1.3438, 1.4609, 0.8945, 0.7852, 1.2422, 1.2422, 1.1016, 1.1875, 1.1641, 1.0703, 1.1016, 1.0547, 1.0703, 0.9102, 1.3828, 1.2188, 1.3438, 1.4219, 1.0078, 1.0312, 1.2109, 1.2344, 0.8438, 1.4297, 0.7422, 1.7891, 0.9414, 1.0625, 0.9102, 1.5078, 0.9453, 1.2344, 1.2188, 0.7695, 0.9531, 0.5039, 0.8047, 1.0391, 1.1953, 1.3828, 1.2812, 1.2578, 1.5156, 0.9844, 0.8633, 0.9453, 1.3828, 1.5234, 1.4453, 1.0234, 1.25, 1.2969, 1.5, 1.2344, 1.5938, 1.0156, 1.0703, 1.3281, 1.1172, 1.1328, 0.9688, 1.1797, 1.1953, 0.8555, 0.9102, 1.2344, 1.5156, 0.9922, 1.3906, 1.2344, 0.7852, 1.0312, 1.4219, 1.3828, 1.3672, 1.1094, 1.5781, 1.6016, 1.0781, 0.9062, 0.8633, 1.7578, 1.0703, 1.375, 0.9961, 1.4844, 0.7305, 1.1328, 0.8906, 1.0156, 1.25, 1.4297, 0.6055, 0.8828, 1.2812, 0.9336, 1.2969, 1.0312, 0.9102, 1.0703, 1.4141, 1.4688, 0.9922, 1.2969, 1.0, 0.8086, 1.2266, 1.4297, 1.1328, 0.625, 1.1641, 0.9375, 1.3594, 1.2188, 0.9375, 1.5781, 1.7344, 1.1172, 0.6641, 1.4766, 0.875, 1.3125, 1.0469, 1.3047, 0.8867, 1.0391, 0.9141, 1.1562, 1.0703, 1.5625, 1.0312, 0.8438, 1.2344, 1.4375, 1.0078, 0.9453, 1.5078, 1.4141, 1.5547, 1.0312, 1.2422, 1.0703, 1.5156, 1.1562, 1.1328, 1.1328, 1.0703, 0.7461, 1.2031, 0.7148, 0.8633, 0.9648, 1.2344, 0.7969, 1.0391, 0.6289, 0.9492, 0.8906, 1.7109, 0.6211, 1.3594, 1.4297, 1.125, 1.7344, 1.7812, 1.4219, 1.6172, 1.2578, 1.3203, 1.0156, 1.2734, 0.7539, 1.4062, 1.2188, 1.6719, 1.7188, 1.2031, 1.8203, 0.832, 1.0547, 1.2812, 1.2734, 1.6016, 1.2344, 1.3906, 1.1328, 1.5469, 1.2734, 1.3594, 1.125, 1.1406, 0.9922, 1.4297, 1.5391, 1.0, 1.0781, 0.957, 1.4062, 1.0234, 1.0938, 1.2188, 0.8594, 1.3125, 0.9258, 1.2109, 1.4062, 1.4688, 1.9453, 1.2031, 0.9336, 0.6289, 1.6172, 1.0078, 1.1094, 1.25, 1.1562, 0.6836, 1.3594, 1.0078, 0.7031, 0.9258, 1.2422, 0.6992, 1.5859, 1.0078, 0.707, 1.4844, 1.4297, 0.9258, 0.9883, 1.5156, 1.3828, 1.1953, 1.1875, 0.8164, 0.9883, 0.957, 1.6328, 1.3438, 0.957, 1.5469, 1.2109, 1.2891, 1.1875, 1.4141, 1.3594, 1.0781, 0.7188, 0.9453, 1.2188, 0.6914, 1.4766, 1.3359, 1.1719, 1.4375, 1.1328, 1.5469, 1.5469, 1.3906, 1.1016, 0.9961, 1.4141, 0.9961, 1.3438, 1.4844, 2.0312, 1.1953, 0.9297, 1.3984, 0.9297, 0.6367, 1.3281, 1.1484, 0.9258, 1.5547, 0.875, 0.8438, 0.9844, 1.7734, 1.1719, 0.832, 1.2266, 1.6094, 0.8242, 0.793, 1.7188, 1.5859, 1.2188, 1.4297, 1.0547, 1.4844, 0.793, 0.9844, 0.8398, 1.0547, 0.8516, 1.0703, 1.5625, 1.2812, 1.6172, 0.7266, 1.0703, 1.5625, 1.3906, 0.8516, 1.3125, 1.125, 1.5078, 0.9141, 0.8555, 1.0547, 0.8359, 0.9766, 1.4219, 1.2422, 1.7969, 1.2188, 1.1562, 1.3125, 1.2422, 1.25, 0.9023, 0.9141, 1.5312, 1.7344, 0.9844, 0.8477, 1.2344, 1.6641, 0.9414, 0.9961, 1.2109, 1.0, 1.3203, 1.2109, 1.0156, 1.6797, 0.9492, 1.5703, 0.9062, 2.125, 1.5703, 1.0859, 1.1953, 0.9531, 1.3359, 0.9375, 1.3047, 1.2422, 1.0078, 1.1875, 1.125, 0.7773, 1.1172, 1.5859, 1.0938, 1.0469, 1.3516, 1.0547, 1.3125, 1.1484, 1.2031, 0.6875, 1.0781, 1.2109, 1.3203, 0.8008, 1.6875, 1.25, 0.668, 1.3828, 1.4375, 0.8828, 1.3984, 1.125, 0.6914, 1.9375, 1.1562, 0.9297, 1.6719, 1.2656, 1.5547, 0.8789, 1.0781, 0.9336, 0.7656, 1.4141, 1.0781, 0.9648, 1.8594, 0.8047, 1.1875, 1.1875, 1.5234, 1.1641, 1.1172, 1.1094, 1.3438, 0.9453, 1.0938, 0.9258, 1.1016, 1.2812, 1.6875, 1.1562, 1.0312, 1.0625, 0.9453, 1.6406, 1.0625, 1.2734, 0.7617, 1.1484, 1.1406, 0.8203, 1.2109, 1.3828, 0.9844, 0.6172, 1.5781, 1.2578, 0.9023, 1.7188, 1.2734, 0.9375, 1.7656, 1.7969, 1.3281, 1.75, 1.3594, 0.8438, 1.1172, 0.9961, 1.2266, 1.0547, 1.4062, 0.9062, 1.25, 1.0938, 1.4922, 1.2266, 0.8945, 1.6328, 1.7656, 0.875, 1.2578, 1.1172, 0.7188, 1.0312, 1.1719, 1.6406, 1.2578, 1.3984, 1.4609, 1.1016, 0.8711, 1.4453, 1.6797, 1.0703, 0.9336, 1.0156, 1.1172, 0.9414, 0.8672, 1.4141, 1.2109, 1.0078, 0.8398, 1.2656, 1.5469, 0.9141, 1.3047, 1.0625, 0.8711, 1.7344, 1.2031, 0.918, 0.8281, 1.0078, 1.4766, 1.4766, 0.8711, 1.2344, 1.1328, 1.3438, 1.0078, 1.3516, 1.0156, 1.3125, 1.3828, 1.0391, 1.2969, 0.6719, 1.0234, 1.0547, 1.2578, 1.0469, 1.125, 1.0156, 1.3047, 1.1328, 1.4609, 0.6719, 1.7734, 1.3672, 1.3672, 1.4297, 1.0234, 1.2578, 0.7031, 1.4922, 1.0234, 1.0391, 1.5625, 1.1719, 1.2188, 1.3359, 1.0547, 1.4688, 1.0312, 1.0625, 1.5859, 1.2422, 1.6641, 1.2734, 1.3281, 1.0938, 1.3906, 1.3984, 0.9492, 1.4297, 0.8672, 1.1562, 1.1797, 0.7266, 1.5781, 1.2188, 1.5781, 1.3594, 1.6719, 1.5859, 1.4062, 1.7969, 1.1797, 0.793, 1.3203, 1.2656, 1.375, 0.9297, 1.7266, 1.1875, 0.9766, 1.125, 1.0859, 1.1484, 1.3281, 1.2188, 0.7148, 1.1094, 1.0469, 0.9531, 0.9336, 1.0, 1.2812, 1.7188, 0.9336, 0.9258, 1.1172, 1.6641, 1.4141, 0.9922, 1.0859, 1.3984, 0.8555, 1.2891, 1.4766, 0.8984, 1.1172, 1.0, 1.0781, 1.5625, 1.4062, 1.1797, 1.3438, 1.0625, 1.5703, 0.6836, 1.1719, 1.1484, 1.2344, 1.1562, 1.5938, 0.9336, 1.2578, 1.1719, 1.2422, 1.3203, 0.8906, 1.2188, 1.25, 0.8281, 0.9922, 1.0078, 1.8359, 1.2344, 0.8164, 1.125, 0.9375, 1.0312, 1.1094, 1.3828, 1.3359, 1.4609, 1.5938, 1.4609, 1.7812, 1.0234, 0.7266, 1.4375, 1.1484, 1.0859, 0.9883, 0.7539, 0.9805, 1.0, 0.6758, 0.7969, 1.0781, 0.8203, 0.7773, 0.9414, 1.5156, 1.2812, 1.4141, 0.8242, 1.5156, 0.6836, 0.6133, 0.9648, 1.5312, 1.125, 0.8867, 1.1719, 1.2109, 1.2734, 0.8359, 0.9453, 1.6328, 1.2188, 1.0781, 1.5469, 1.4453, 1.1797, 1.0469, 1.7109, 0.8828, 0.8594, 1.3047, 0.9297, 1.1641, 1.375, 0.8086, 1.1719, 0.9922, 1.4062, 1.125, 1.6797, 1.1406, 1.125, 1.3047, 1.2266, 1.1328, 1.1719, 1.5156, 1.1641, 1.1328, 1.2969, 1.4766, 1.3516, 0.7266, 0.6719, 1.3203, 1.3594, 1.1875, 1.2266, 1.0859, 1.2656, 1.4141, 1.0234, 0.8906, 1.4609, 0.8477, 1.2422, 1.1797, 1.1328, 1.0859, 1.3203, 0.8398, 0.7969, 0.9844, 1.1328, 1.3125, 1.3906, 1.0391, 0.9258, 1.2891, 1.5312, 1.3594, 1.3828, 1.5, 0.8281, 1.4609, 1.1094, 1.1953, 1.0547, 1.375, 1.1094, 0.9844, 0.7695, 1.5, 1.5469, 0.9922, 1.2266, 0.8477, 1.1328, 1.1641, 0.9375, 1.1641, 1.4766, 1.0312, 1.1328, 1.5078, 1.6719, 1.4531, 1.4922, 1.0859, 1.9766, 1.3984, 1.4844, 1.3359, 1.3047, 1.3203, 1.1641, 0.8125, 1.1719, 1.0312, 1.2969, 1.0547, 1.2578, 1.2656, 0.957, 0.7461, 0.8008, 1.4141, 1.1016, 0.7461, 1.4922, 0.793, 1.2031, 1.2578, 1.125, 1.0, 0.8164, 1.1406, 0.9375, 1.2188, 1.2266, 1.0547, 1.25, 0.9727, 1.2969, 1.0312, 0.9023, 1.5234, 1.4141, 1.8438, 1.4688, 1.3828, 1.4062, 1.1172, 1.0078, 1.0391, 1.2891, 1.4219, 0.8047, 1.6328, 1.0625, 1.1562, 0.6484, 0.7305, 0.9609, 1.5312, 1.1719, 0.9531, 1.0312, 1.1484, 0.9297, 1.0547, 1.5312, 1.1172, 1.3516, 0.8281, 0.9297, 0.8633, 1.3281, 1.6875, 1.0625, 1.2344, 1.3984, 0.9531, 1.1172, 1.1094, 1.1875, 0.9648, 1.2578, 1.1328, 1.0938, 1.4375, 0.918, 0.6211, 1.375, 1.2656, 0.9922, 0.8203, 1.0781, 0.9883, 0.7891, 1.1172, 1.0391, 1.0078, 1.5625, 0.9766, 1.3984, 0.9648, 1.0781, 1.1797, 1.0312, 1.25, 1.6172, 2.2812, 1.2188, 0.8477, 1.1328, 1.4922, 1.6016, 1.0156, 1.6953, 1.0234, 1.6172, 1.0781, 0.9297, 1.3281, 1.0469, 1.0781, 1.3047, 0.9102, 1.5, 1.5781, 0.9102, 1.2891, 0.7031, 0.9375, 0.6719, 1.1875, 0.7148, 0.8984, 0.6289, 1.6641, 1.2578, 1.4609, 1.0703, 1.0391, 1.4766, 1.1875, 0.9609, 0.9844, 0.9062, 0.8086, 0.8906, 1.0078, 1.6719, 1.6094, 1.2266, 1.125, 1.6406, 1.4219, 0.7656, 0.957, 1.7969, 1.0859, 1.0859, 1.5234, 1.0312, 1.0234, 1.0156, 1.1719, 0.9688, 1.1484, 1.1875, 0.9609, 1.0625, 1.0938, 1.0781, 1.5, 1.1641, 1.5312, 1.0078, 0.9023, 1.1406, 1.1562, 1.0, 1.3438, 1.2266, 0.7539, 0.8516, 0.8867, 0.793, 0.8906, 0.8945, 0.8281, 1.0781, 1.4531, 0.8203, 1.5234, 1.4375, 1.0859, 1.1094, 1.0234, 0.6953, 1.1562, 1.3516, 1.1562, 0.7969, 1.0859, 1.5078, 0.9297, 1.3984, 1.3438, 1.5781, 1.1719, 1.0078, 0.9531, 1.125, 1.0859, 1.2188, 1.3203, 1.1875, 1.4219, 1.6094, 1.75, 1.2812, 0.8945, 0.7461, 1.1172, 1.5234, 1.3438, 0.8633, 0.9883, 1.8516, 1.0156, 1.0, 1.1719, 1.1641, 1.6172, 1.3047, 0.9766, 1.3516, 0.707, 1.3594, 0.9922, 1.2734, 0.9336, 1.1484, 1.4141, 0.9883, 1.7109, 1.5547, 0.6875, 0.8828, 1.1328, 1.2969, 0.9492, 0.8125, 1.1641, 1.3047, 0.9805, 1.125, 1.3125, 1.0781, 0.9727, 0.707, 1.1016, 1.1797, 1.3438, 1.0547, 1.4453, 1.2891, 0.5508, 0.6953, 1.1172, 1.1641, 1.3672, 1.0938, 1.3203, 0.7617, 1.0625, 1.2656, 0.8633, 0.8047, 1.1094, 1.2188, 0.9883, 1.6719, 1.4766, 0.8711, 1.3438, 1.3672, 0.8867, 1.6328, 1.2109, 0.9023, 1.4609, 1.2109, 0.875, 1.3594, 1.3047, 0.8164, 1.5312, 0.9297, 1.4062, 1.2891, 1.1016, 0.7266, 1.0391, 1.3828, 1.2656, 1.1484, 0.8164, 0.8242, 1.1719, 0.7031, 0.9492, 0.9297, 1.2266, 0.957, 1.1328, 1.2109, 1.2734, 1.5625, 0.8945, 1.7891, 0.9922, 1.0156, 1.2344, 0.9336, 1.4531, 1.9297, 1.2031, 0.9297, 1.4219, 0.9648, 1.3047, 1.1953, 1.2188, 0.7148, 1.2969, 1.5391, 0.8047, 0.875, 1.2031, 1.1328, 1.4219, 1.4453, 1.0703, 1.0, 1.2422, 1.5625, 1.4531, 1.6719, 1.0938, 0.9102, 1.5703, 0.8164, 1.2656, 0.8242, 1.2422, 1.6562, 1.4766, 0.7539, 1.125, 1.0547, 1.3359, 1.1719, 1.2422, 2.0781, 1.0547, 1.6797, 1.2656, 1.1328, 0.7344, 1.0859, 0.5859, 1.1875, 1.3281, 1.1328, 1.4453, 1.6484, 0.7812, 1.6016, 1.2578, 1.3203, 1.25, 1.2188, 0.8672, 1.5234, 1.2031, 1.1719, 1.0469, 0.75, 1.8828, 1.5234, 0.832, 1.2656, 1.0078, 1.0234, 0.7266, 1.0625, 1.25, 1.3906, 0.832, 0.6758, 1.7734, 1.4609, 0.9883, 0.9805, 1.1172, 1.4531, 1.1328, 1.9141, 1.7578, 0.9648, 1.3828, 1.0703, 0.8828, 1.2188, 1.2969, 0.8242, 1.1016, 0.9805, 0.9336, 1.4844, 0.7344, 0.9414, 1.8906, 1.3516, 1.0078, 0.8008, 0.9609, 1.3984, 0.7695, 1.3828, 1.5078, 1.2656, 1.0859, 1.5, 1.4297, 1.9297, 1.1406, 1.3438, 1.1406, 0.9531, 1.1172, 1.4297, 0.9961, 1.2422, 0.9023, 1.2734, 1.0469, 0.8594, 1.0156, 1.3828, 1.7891, 1.2344, 1.5156, 1.4219, 0.9102, 1.75, 0.8438, 1.2578, 1.4844, 0.9766, 1.0, 1.2266, 1.2344, 1.3203, 0.7656, 2.5156, 0.6953, 1.6641, 1.0625, 1.6562, 1.0, 0.7852, 1.0703, 0.9922, 1.3047, 0.8984, 1.2578, 0.9805, 1.1406, 0.7734, 1.3906, 1.3125, 1.125, 1.0312, 0.5547, 1.2656, 1.2109, 0.9336, 1.0312, 1.4062, 0.9336, 0.8984, 1.2109, 1.4609, 1.1797, 1.0312, 1.4375, 0.832, 1.1719, 0.9531, 1.0391, 1.0469, 1.2188, 0.8867, 1.3906, 1.5391, 1.0938, 0.6289, 0.7891, 0.9492, 0.5312, 1.0781, 0.8242, 0.9023, 1.25, 1.3281, 1.3438, 1.3516, 1.2422, 0.7461, 1.0391, 1.5234, 1.1641, 0.7148, 1.1953, 1.1172, 1.3594, 1.1094, 1.1406, 1.1719, 1.2578, 1.0703, 0.8164, 1.5859, 1.2422, 0.8086, 1.0234, 1.4922, 1.1797, 0.9297, 0.9844, 1.0625, 1.0469, 0.9961, 1.2812, 1.0703, 2.0312, 0.7891, 0.9453, 1.1797, 1.0312, 1.0078, 0.7539, 1.125, 0.7422, 1.1172, 1.2891, 1.2656, 1.5781, 0.8984, 0.9844, 1.0156, 1.2812, 0.8281, 1.75, 0.9727, 1.0, 1.1484, 0.9688, 1.1406, 1.1328, 1.4766, 1.0, 1.4219, 1.0391, 1.3125, 1.0391, 1.0781, 1.25, 1.3984, 1.9609, 1.4062, 1.5156, 1.2812, 1.1172, 1.1172, 1.3516, 0.9414, 1.3125, 0.957, 0.6562, 1.1094, 0.9219, 1.6016, 1.3438, 1.0078, 1.1719, 1.5, 1.1406, 1.7422, 1.1172, 0.8984, 1.0625, 0.6758, 0.8867, 1.0469, 0.582, 0.9492, 1.2891, 0.8164, 1.6562, 0.9609, 0.832, 1.2812, 1.2109, 1.4297, 1.3672, 1.2109, 1.0547, 1.1875, 1.0938, 1.8359, 1.5625, 1.1484, 1.0156, 1.375, 1.2422, 1.1094, 1.6953, 1.1641, 1.0859, 1.1641, 1.0703, 0.9727, 1.1719, 1.4297, 0.9219, 1.2656, 1.2891, 1.1328, 0.8828, 0.9844, 0.8789, 1.0547, 1.2734, 2.0312, 1.4141, 1.1562, 1.4688, 1.125, 1.1562, 1.1172, 1.3203, 1.2031, 0.8906, 0.7148, 0.918, 0.957, 1.1719, 1.5469, 1.3672, 1.2109, 1.3516, 0.8281, 1.3594, 1.0156, 1.1562, 1.2656, 1.0547, 1.0156, 0.6797, 1.4688, 1.375, 1.5781, 0.9883, 1.4453, 0.9414, 0.7383, 1.3125, 1.6406, 1.4062, 0.7773, 1.0625, 0.6992, 0.7422, 1.2734, 1.0234, 1.0781, 1.2656, 0.9492, 1.7109, 1.0938, 1.0156, 1.7422, 1.2812, 1.3828, 1.375, 0.6719, 1.4219, 0.7812, 0.9727, 0.7656, 0.707, 1.1797, 1.8047, 1.1875, 1.2812, 1.2031, 0.7344, 0.957, 0.8555, 1.6406, 1.1016, 1.0312, 1.9766, 1.0, 1.0859, 1.1719, 1.0938, 1.7266, 1.1016, 1.5859, 1.4219, 0.8633, 1.2422, 0.9922, 1.1641, 1.9062, 1.0078, 1.1328, 1.0469, 0.5859, 1.1406, 1.0078, 0.5664, 1.0156, 0.8086, 1.6719, 0.9141, 1.1953, 1.1719, 1.5156, 1.3516, 0.9727, 1.4609, 1.1406, 1.2734, 0.8047, 1.4219, 0.8398, 0.75, 1.75, 0.8711, 0.9492, 1.2031, 1.1328, 0.8203, 0.6875, 1.7422, 0.7617, 0.8906, 1.0078, 1.2344, 1.1953, 1.2188, 1.0, 0.9336, 1.2344, 1.2891, 0.8281, 1.2031, 1.0156, 1.1641, 0.7227, 0.7969, 1.0, 1.1016, 1.1172, 0.875, 1.5547, 0.8125, 1.1094, 0.9492, 1.4219, 1.4375, 1.8516, 1.1562, 1.3906, 1.0703, 1.1641, 1.0625, 1.2344, 1.3047, 1.1172, 1.2422, 1.3047, 1.6875, 1.4453, 0.7773, 1.5859, 0.7305, 1.1016, 1.4219, 1.0781, 0.7305, 1.3672, 1.3438, 1.2266, 1.0703, 1.2734, 1.5547, 1.125, 1.5469, 1.25, 1.0625, 0.9766, 0.875, 1.3281, 1.3828, 1.0312, 1.0859, 1.1562, 1.3594, 1.0859, 1.4766, 1.2891, 1.0625, 0.8984, 1.2812, 0.8594, 1.1953, 0.8711, 1.1641, 1.0703, 1.0, 1.3125, 0.9297, 0.7734, 1.3984, 1.3203, 1.0469, 1.3516, 1.125, 0.8164, 0.9688, 0.9258, 0.9883, 0.7305, 0.832, 1.2734, 1.3828, 0.7695, 0.8086, 0.7266, 0.9531, 1.2422, 0.8359, 0.8633, 1.3672, 1.0938, 1.1562, 1.0078, 0.9453, 1.0, 1.0469, 1.3438, 1.2188, 1.2031, 1.75, 1.0234, 1.0469, 0.6875, 1.1328, 0.9453, 1.2578, 0.8438, 1.5469, 1.8281, 1.5234, 0.9375, 1.4141, 1.5078, 0.9688, 0.9375, 1.6328, 0.8945, 0.8672, 1.7266, 1.1875, 0.7852, 1.6172, 0.7227, 1.7891, 1.3828, 0.8984, 0.9414, 0.9258, 0.8945, 1.2734, 0.7422, 1.0156, 0.9375, 1.1094, 1.2109, 0.9492, 1.5781, 1.0625, 0.7266, 1.1797, 0.9609, 1.3906, 1.3438, 1.4609, 0.9766, 1.2656, 1.1484, 1.2812, 1.7344, 0.9062, 1.0703, 1.4453, 1.5938, 1.1641, 1.2891, 1.0156, 1.5625, 0.7656, 1.7812, 0.9805, 0.8555, 0.8984, 1.1953, 1.0547, 1.2969, 1.3594, 0.9336, 1.4922, 0.8359, 0.8672, 1.2578, 1.1328, 1.4219, 0.9414, 1.6797, 0.8359, 0.9141, 0.9844, 0.9648, 1.6172, 1.2812, 0.9883, 1.3281, 1.2969, 1.2109, 1.0859, 1.5859, 1.6328, 0.793, 1.1797, 1.2188, 0.9219, 0.6523, 1.1641, 1.4375, 0.8945, 0.9102, 1.3594, 1.3203, 0.9648, 1.3203, 1.0781, 0.668, 1.2266, 1.1641, 1.1797, 1.3438, 1.0859, 1.3594, 1.2656, 1.4531, 0.9805, 0.9375, 0.875, 1.2656, 1.0469, 1.3125, 0.9961, 1.6641, 0.9219, 1.3281, 1.3516, 0.8398, 0.8281, 1.2422, 1.3672, 1.25, 1.3359, 1.0547, 1.3672, 1.1797, 1.3672, 1.2812, 1.2969, 1.7812, 0.8438, 0.9375, 1.2891, 1.6406, 0.7734, 1.0391, 1.0781, 1.2578, 1.1094, 1.0547, 1.1641, 1.1406, 0.8438, 1.3906, 1.2188, 0.9883, 1.5547, 1.4375, 1.7656, 1.4219, 1.3984, 1.1797, 0.9688, 0.7461, 0.8047, 1.2969, 0.7969, 1.0469, 1.0469, 1.0234, 0.7188, 0.707, 1.0625, 0.8789, 1.3125, 1.0625, 1.0781, 1.4688, 1.2266, 1.0547, 1.2109, 1.6875, 0.7109, 1.5, 1.5703, 0.6836, 1.4922, 1.3594, 1.2344, 1.0781, 1.1719, 0.7695, 1.2109, 1.2891, 1.2344, 1.3516, 1.0156, 1.2969, 1.2031, 0.7578, 1.0078, 1.1641, 0.7578, 1.0078, 1.1719, 1.1172, 1.3359, 1.0391, 1.3594, 1.9375, 0.9492, 0.9297, 0.6641, 0.9453, 1.0234, 1.0859, 1.0156, 0.9453, 1.3359, 1.3984, 1.0, 0.7891, 0.832, 1.9688, 1.1719, 0.9297, 1.1953, 1.1406, 1.2891, 1.1484, 0.7148, 1.7109, 1.5781, 1.7422, 1.0547, 1.6797, 0.8789, 0.9375, 0.9062, 1.2109, 1.5859, 0.9492, 0.8555, 0.8516, 1.1484, 1.4297, 1.2266, 1.2656, 0.7891, 1.3047, 0.9375, 1.1562, 0.7969, 1.5, 0.8008, 0.8516, 0.9648, 1.2344, 0.9648, 1.3438, 1.3438, 1.2578, 1.1016, 1.4297, 1.2891, 0.8477, 0.8008, 1.7109, 1.7344, 1.4141, 1.1953, 1.5234, 1.5938, 1.4141, 0.9727, 0.7344, 1.125, 0.5547, 1.2891, 0.8086, 1.1641, 1.1719, 1.3125, 1.5547, 0.9336, 1.4609, 1.4297, 1.2812, 1.5703, 1.125, 1.3047, 1.2812, 0.9258, 0.6602, 1.3438, 1.0156, 1.3438, 1.4062, 1.3438, 0.9609, 1.2969, 0.7969, 1.2812, 1.3594, 0.9453, 1.0, 0.9492, 1.4141, 0.9727, 1.8203, 0.832, 1.1953, 1.4844, 1.1562, 1.3125, 1.1875, 0.7227, 1.3281, 1.1797, 1.4062, 0.9883, 0.7969, 1.4297, 0.8008, 0.6367, 1.6719, 0.9453, 0.7266, 2.0781, 1.2344, 1.5781, 1.1797, 1.4844, 1.1875, 1.0859, 1.0078, 1.2266, 1.2109, 1.375, 1.3438, 1.4375, 1.0156, 1.6562, 0.9531, 1.5469, 0.7344, 0.9414, 1.1328, 1.0859, 0.9688, 1.125, 1.4844, 1.4531, 1.0703, 1.0547, 0.9922, 1.2266, 1.125, 1.4297, 1.7656, 0.832, 0.9805, 1.4688, 1.1094, 1.4688, 0.918, 1.1484, 0.9648, 0.918, 0.9648, 0.7031, 0.9453, 0.6914, 1.1406, 0.8047, 0.8555, 1.5312, 1.3984, 1.1641, 0.7578, 1.1875, 0.9297, 0.9688, 1.0078, 1.2109, 0.6719, 1.1641, 0.9102, 0.8164, 1.0859, 1.1328, 1.5, 0.875, 1.2734, 1.3516, 1.1172, 1.4922, 1.7891, 1.5, 0.8555, 1.0938, 1.4844, 0.5898, 1.6719, 0.8555, 1.2891, 1.2109, 1.3984, 0.7344, 1.2578, 1.3906, 1.4688, 0.9492, 1.1016, 1.4062, 0.9414, 0.957, 0.8594, 1.0625, 0.9414, 1.8672, 0.7461, 1.2812, 0.918, 1.2344, 0.9258, 1.4297, 1.7344, 1.2578, 0.7852, 1.2031, 1.0469, 0.8047, 1.7734, 0.7969, 1.0938, 0.9766, 0.8672, 1.3672, 0.9336, 1.2969, 0.9609, 1.6875, 1.6562, 0.7109, 1.3984, 1.3125, 1.0703, 1.8438, 1.0078, 1.3438, 1.2188, 0.5273, 0.8203, 1.3516, 1.3672, 1.4297, 1.2422, 1.1641, 1.1172, 0.9805, 2.1562, 0.9297, 0.7773, 0.8984, 0.8672, 1.6406, 0.9062, 1.0703, 1.4062, 1.3984, 0.9609, 1.4297, 0.6562, 1.1016, 0.9219, 0.7617, 1.6172, 0.8438, 1.25, 1.3516, 0.7109, 1.1484, 0.5859, 1.375, 0.8594, 1.2031, 1.0625, 0.9141, 0.7422, 1.0859, 1.5078, 0.7734, 1.5469, 0.7539, 0.8164, 1.5, 1.0938, 1.2344, 1.5938, 1.375, 1.2422, 1.6094, 0.875, 1.3438, 1.6953, 1.2188, 1.1953, 0.8398, 1.5078, 1.0547, 1.3672, 0.7773, 0.9961, 0.9844, 1.2969, 1.3125, 1.1797, 1.1719, 1.0469, 1.3672, 0.9141, 1.2031, 1.75, 1.4375, 0.9414, 0.9141, 1.1328, 1.0703, 1.5391, 1.1484, 1.1484, 0.9297, 0.9727, 1.4062, 0.7773, 1.2812, 1.1562, 0.9844, 1.0703, 1.1719, 1.2969, 1.0938, 0.8867, 1.7812, 1.3438, 0.625, 1.1719, 0.875, 0.9531, 1.4766, 1.2344, 1.5547, 0.6133, 0.7695, 1.0547, 1.1406, 1.5, 1.3359, 1.9062, 1.0781, 0.9102, 1.0938, 0.918, 1.3125, 1.0625, 1.25, 1.1641, 1.7344, 1.2422, 1.2422, 1.4297, 1.3906, 1.7422, 1.2734, 1.3516, 1.1719, 0.7969, 1.3906, 1.8281, 1.1094, 1.2344, 0.6992, 1.1484, 1.4844, 1.0938, 0.8867, 1.3203, 1.1016, 1.2188, 1.3125, 1.1719, 1.4453, 1.1641, 1.25, 1.2656, 1.2734, 0.9258, 1.2812, 1.5078, 1.625, 0.9492, 0.9375, 1.0078, 1.2188, 1.1328, 1.1719, 0.668, 1.1641, 1.0547, 0.8242, 1.1406, 1.4766, 1.0391, 1.5859, 1.3594, 1.1094, 0.8516, 1.2734, 1.3672, 1.2344, 1.6953, 0.9805, 0.8867, 0.6836, 1.0469, 1.625, 1.4844, 1.5859, 1.2422, 1.0625, 0.7617, 1.2656, 0.9727, 0.918, 0.8789, 0.7227, 1.3359, 1.2969, 1.4219, 1.1094, 1.6641, 1.2109, 1.3984, 1.1953, 1.1016, 1.2422, 1.3594, 0.6133, 1.3281, 0.9453, 0.5977, 1.25, 1.0625, 0.7773, 1.3359, 1.0547, 0.9219, 1.6797, 1.3281, 1.0781, 1.2891, 0.8867, 1.4531, 0.7422, 0.9336, 0.9414, 1.2109, 0.7383, 1.1719, 1.6172, 0.9688, 1.3438, 1.0859, 1.1953, 1.5078, 1.1094, 0.7578, 1.4453, 0.9141, 1.0234, 1.3906, 1.4375, 1.0156, 0.9609, 1.0547, 0.7852, 1.2969, 1.4062, 0.9531, 0.8477, 0.9336, 1.1484, 1.1875, 1.1719, 1.0078, 0.6367, 0.7969, 1.0078, 1.5469, 1.5234, 0.875, 0.9648, 0.6758, 1.6406, 1.4297, 1.6641, 1.2344, 1.7188, 1.2266, 0.6133, 1.1016, 0.7109, 1.1875, 0.6602, 1.3359, 0.8672, 0.8906, 0.9062, 1.0234, 1.0312, 0.8828, 1.0312, 1.1875, 0.832, 1.2109, 0.9023, 1.1406, 1.5625, 1.0469, 0.8164, 1.2734, 1.0156, 1.6562, 1.2578, 1.3359, 1.4219, 1.6094, 0.9922, 0.9375, 0.832, 1.3203, 1.1641, 1.3984, 0.6367, 1.3828, 1.0, 0.9414, 1.3281, 0.8711, 1.2188, 1.7344, 0.875, 0.8086, 1.1094, 0.6797, 0.625, 1.6484, 0.9883, 1.2266, 1.8906, 1.0234, 1.0547, 0.8672, 0.6797, 1.3203, 0.8008, 1.6172, 0.9492, 1.4609, 1.5312, 1.2656, 1.7734, 1.1172, 0.9219, 0.9609, 1.3047, 1.2422, 0.8828, 1.6641, 1.2031, 0.9453, 0.918, 1.4062, 1.4844, 1.0, 1.3047, 1.0938, 0.7305, 0.7578, 1.2031, 1.4062, 0.9219, 1.3984, 1.1953, 1.1875, 1.2891, 1.125, 1.0625, 1.0938, 0.9805, 0.8594, 1.1094, 0.9961, 0.9648, 0.7227, 0.7969, 1.8125, 0.8594, 1.2734, 1.2578, 0.832, 1.0547, 1.5469, 0.8711, 1.3984, 0.9102, 1.0469, 0.7461, 1.6094, 0.8555, 1.4219, 0.9219, 1.1797, 0.8047, 1.1719, 1.0625, 1.0234, 0.9023, 1.375, 1.0859, 1.0547, 1.0859, 1.1953, 1.1406, 1.2422, 1.1016, 1.3359, 1.2266, 0.7891, 0.7852, 0.7734, 1.8438, 0.9219, 1.0234, 1.4844, 1.25, 0.7305, 1.1016, 1.4844, 1.0547, 1.5781, 0.9922, 1.1719, 1.5547, 1.25, 0.5781, 0.7266, 0.6289, 1.3281, 2.0469, 1.3125, 1.6484, 1.6172, 0.7695, 1.2188, 0.9766, 1.2266, 0.9609, 1.0078, 0.7812, 1.3125, 1.3281, 1.0, 1.1406, 1.1016, 1.0078, 1.1797, 1.0078, 1.2344, 1.2109, 1.0312, 0.7773, 0.7773, 1.2031, 0.9844, 0.6836, 0.9297, 0.9805, 1.5156, 1.0469, 0.9688, 1.3828, 1.2188, 1.0234, 1.1953, 1.0703, 0.6602, 1.0469, 0.8555, 1.375, 0.9258, 1.2188, 0.9961, 0.9648, 1.1172, 1.4609, 1.125, 1.25, 0.7617, 1.3594, 1.1406, 0.8359, 0.8711, 0.9492, 0.707, 1.4219, 1.6562, 0.9648, 1.1172, 1.0859, 0.7969, 0.8828, 1.1719, 0.7656, 0.6953, 0.7148, 1.0234, 1.5938, 1.4922, 1.4609, 0.9258, 1.4297, 1.3438, 1.1562, 1.3984, 1.1797, 1.4531, 1.5156, 1.2734, 1.4062, 1.7031, 1.4766, 1.0234, 1.2188, 1.7266, 1.3906, 0.9375, 0.8516, 1.0781, 0.875, 1.2578, 1.7109, 0.9375, 1.2031, 1.4453, 0.9766, 1.3516, 1.6016, 0.5859, 1.2266, 1.3047, 1.0078, 0.7773, 0.668, 1.0469, 1.4609, 0.9961, 0.8398, 1.1797, 0.9336, 1.0156, 0.9844, 1.0469, 0.9844, 0.9297, 1.1562, 1.1406, 1.4922, 1.125, 1.1641, 1.9453, 1.2344, 1.8828, 0.9414, 1.0859, 0.8164, 1.0078, 1.1875, 1.5547, 0.8164, 1.4453, 0.582, 1.3672, 1.4219, 1.1172, 1.4453, 0.8164, 1.0625, 0.8086, 1.2344, 1.3047, 1.5312, 1.1016, 1.3594, 1.5547, 1.1016, 1.125, 1.6094, 1.125, 0.7305, 1.1641, 0.9023, 1.1484, 0.8828, 0.9414, 1.0234, 1.0469, 1.1797, 1.0547, 0.7812, 1.7109, 1.0312, 1.0234, 1.3984, 0.8867, 1.0625, 1.3984, 1.0391, 0.8203, 1.1172, 1.0625, 1.0, 1.2188, 1.25, 1.0703, 1.1094, 1.1719, 0.9766, 1.0625, 1.3516, 0.8359, 1.3203, 1.0391, 1.4219, 1.5078, 1.4922, 1.4844, 0.7031, 1.4297, 1.0547, 1.0859, 0.9141, 0.6914, 1.125, 1.2109, 0.8867, 1.1172, 1.5391, 1.5625, 1.7188, 0.9258, 1.1406, 1.4453, 1.4609, 1.6172, 1.5625, 1.3984, 1.375, 1.4922, 1.3281, 1.8438, 0.9805, 1.0625, 1.5703, 0.9492, 0.7422, 2.4531, 1.375, 1.125, 1.0312, 1.0547, 1.1172, 1.7109, 2.1094, 1.0703, 1.2969, 1.1172, 1.5938, 1.5859, 1.3828, 0.9453, 1.2656, 1.0156, 1.8203, 0.9258, 1.0312, 1.4688, 0.9102, 0.9883, 1.4375, 0.8672, 1.1484, 1.4609, 1.5625, 0.6875, 0.8633, 0.75, 1.4062, 0.9023, 1.3906, 1.1328, 1.125, 1.2656, 0.9805, 1.0234, 1.5234, 1.3359, 1.1016, 1.0234, 0.707, 1.2422, 0.9727, 1.4531, 1.3594, 0.7461, 1.4922, 1.0234, 1.2812, 1.7656, 0.9336, 1.4844, 0.8516, 1.1016, 1.5234, 0.9531, 0.8164, 0.8555, 1.4219, 0.7773, 1.0625, 1.3281, 1.4297, 1.0156, 1.4062, 1.1797, 0.9141, 1.2422, 0.9727, 0.8867, 0.9961, 0.9062, 0.9805, 0.9961, 1.8516, 1.1016, 0.9102, 0.6406, 0.7891, 1.0547, 0.7422, 1.1406, 1.1484, 1.3281, 1.2109, 0.9258, 1.1172, 0.8828, 0.8594, 0.9688, 0.8086, 1.2656, 1.3203, 1.2891, 0.9492, 0.7461, 1.0234, 0.9648, 1.5312, 1.0703, 1.9844, 1.1562, 1.0156, 1.2578, 1.5312, 1.1016, 1.4062, 1.0234, 0.9609, 1.0625, 1.2812, 1.4141, 1.0859, 1.25, 1.2109, 1.375, 1.3125, 0.8828, 0.8672, 1.1953, 1.0469, 1.2188, 0.5781, 1.9453, 1.0859, 1.2266, 1.0938, 0.7031, 1.5391, 0.8359, 0.9023, 0.7812, 1.5312, 1.4609, 0.9531, 0.6602, 1.4766, 1.3516, 1.0703, 0.9492, 0.7031, 1.8516, 1.4141, 1.0156, 1.0859, 1.1406, 0.9531, 1.3672, 0.75, 1.1172, 1.5, 1.6875, 0.8594, 1.0703, 1.0938, 1.1484, 1.4609, 1.0859, 1.2656, 0.7812, 1.2109, 1.5078, 1.0938, 0.9844, 1.3516, 1.2578, 1.0, 0.6094, 1.0469, 0.9961, 1.0938, 0.8398, 1.1484, 0.8984, 1.0469, 1.4766, 1.4141, 1.5547, 0.918, 0.6289, 1.2266, 1.0, 1.375, 1.2188, 1.6641, 0.7969, 1.1172, 1.0312, 0.7539, 0.6875, 1.6484, 1.2969, 1.0547, 1.8828, 0.7773, 1.8281, 1.5938, 1.1953, 1.5234, 1.375, 1.2578, 1.2266, 1.375, 1.0781, 1.5, 1.0781, 1.0156, 0.9922, 1.3984, 1.6562, 0.8789, 2.1875, 0.7812, 0.9961, 1.5078, 0.8633, 0.8477, 1.0156, 0.75, 1.2422, 1.1328, 0.7266, 1.1172, 0.7812, 0.7344, 0.8516, 1.1094, 1.1484, 1.5078, 0.7891, 1.2812, 1.4453, 1.2891, 0.9922, 1.6797, 1.4141, 1.0, 0.6016, 1.3906, 0.7617, 1.2188, 2.0156, 0.9297, 1.2422, 1.375, 0.9609, 1.2109, 1.2031, 1.0469, 0.9297, 0.8633, 0.9023, 0.8984, 0.8711, 1.1562, 1.2578, 1.2891, 1.4688, 1.6484, 1.2188, 1.3125, 0.9648, 1.1094, 1.5391, 1.0703, 1.1562, 1.0703, 0.9062, 1.1094, 1.4531, 1.3281, 1.2344, 0.6445, 0.9141, 1.4609, 1.6641, 1.1406, 0.8516, 1.8516, 0.7695]
|
Chapter 3/chapter_3_loss.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
[1.9531, 1.3281, 4.0938, 4.8125, 3.7344, 2.7812, 3.4219, 2.7656, 2.9531, 5.75, 3.1562, 3.0469, 4.6562, 3.9844, 3.7969, 2.4375, 2.3125, 4.75, 2.4531, 2.8125, 3.8594, 2.25, 1.3281, 2.6094, 3.0938, 1.2344, 3.7188, 2.3281, 2.1406, 2.7031, 2.7656, 2.8594, 2.7969, 2.9688, 3.6094, 2.0781, 3.5938, 2.5, 1.8906, 2.0938, 2.4688, 2.7656, 2.3594, 5.5625, 3.4219, 2.9062, 1.0625, 1.5156, 2.2812, 3.3438, 3.8906, 2.8906, 4.125, 3.9688, 0.8398, 1.7344, 3.7031, 3.75, 2.5156, 2.75, 2.8594, 1.9062, 1.2344, 1.625, 1.4844, 2.6562, 2.3594, 1.2578, 1.9531, 2.2344, 2.4844, 3.7656, 3.1094, 1.75, 3.3125, 2.5, 3.0469, 1.7109, 1.0391, 3.6719, 2.8281, 1.3203, 0.8164, 2.1094, 2.5156, 2.7656, 2.9375, 3.0156, 3.0312, 3.9375, 2.7188, 0.9219, 1.5938, 2.875, 3.5938, 1.9297, 3.4844, 2.6406, 1.0781, 1.9766, 1.6172, 2.8438, 2.2031, 3.4531, 0.9453, 3.3438, 2.5625, 2.7344, 2.75, 4.8438, 2.75, 3.5, 2.3438, 2.1094, 2.4375, 3.375, 2.7031, 3.8281, 2.5469, 2.8594, 1.2344, 1.625, 3.9844, 3.9531, 3.9062, 1.0156, 3.2188, 1.2109, 2.6094, 3.7188, 2.5938, 4.5, 2.6719, 1.4844, 3.5469, 4.5625, 2.6875, 1.5156, 2.5156, 2.6406, 2.1094, 3.2969, 2.0156, 1.9219, 3.9062, 2.5156, 3.125, 2.2188, 2.4688, 0.8438, 3.5469, 2.6094, 2.7188, 1.4219, 2.75, 0.9805, 2.5938, 2.7812, 1.7344, 2.1406, 2.75, 2.8906, 2.3438, 4.7188, 3.25, 1.6484, 3.3438, 3.3594, 0.8555, 3.5625, 2.625, 2.0625, 3.9531, 2.3906, 4.625, 3.8906, 1.7188, 2.7031, 3.4219, 3.3438, 3.8594, 1.6406, 0.8203, 3.0312, 5.7188, 2.3906, 1.6328, 3.2031, 2.7031, 1.6016, 3.5312, 3.3594, 3.3438, 1.2656, 0.8164, 2.1406, 2.0938, 0.8359, 2.7188, 2.7969, 2.3281, 2.4531, 2.2969, 2.6719, 2.7344, 2.7969, 3.0312, 2.4688, 0.8633, 2.6875, 1.625, 2.4375, 2.7188, 3.0781, 4.3438, 2.8438, 5.2812, 0.8633, 2.4375, 2.5781, 1.8203, 3.6094, 1.9062, 1.0, 2.7812, 1.6953, 2.6875, 3.1562, 2.3906, 3.1562, 2.25, 2.5469, 0.8047, 3.2656, 2.4844, 2.25, 2.5938, 3.4062, 2.5938, 2.4531, 1.1562, 0.6367, 2.8906, 2.6094, 3.0, 4.0, 5.0938, 2.0312, 3.2656, 2.2656, 1.8359, 4.4688, 1.9219, 3.4531, 2.7656, 2.2188, 2.8594, 3.0312, 1.5391, 2.6875, 2.2344, 3.8594, 2.4531, 2.6406, 1.7422, 3.0, 2.9219, 1.4531, 2.4062, 1.5703, 1.8594, 1.6797, 2.6094, 1.1406, 2.3281, 1.3047, 3.5312, 2.5938, 4.8125, 2.9062, 2.3906, 1.0078, 4.6875, 2.2031, 1.375, 0.8438, 1.4922, 0.8711, 1.3672, 0.832, 2.8438, 3.0156, 2.7812, 3.2969, 2.9531, 2.1094, 0.8047, 2.7344, 2.8281, 2.0938, 2.3906, 3.0, 1.0234, 0.6406, 2.9219, 2.5312, 1.8984, 2.8125, 2.8594, 2.9062, 1.8984, 2.3906, 2.8906, 0.8086, 4.2812, 3.1094, 2.0469, 2.6094, 1.9844, 4.5312, 2.1875, 0.8828, 0.9492, 2.25, 2.0938, 3.0312, 1.6719, 2.125, 2.6719, 1.9531, 2.5781, 2.6875, 2.6875, 0.7578, 2.7188, 0.8633, 3.0469, 2.2812, 4.125, 2.9062, 2.7969, 1.5625, 1.8359, 1.0703, 1.8125, 2.8281, 3.4062, 2.6875, 0.7617, 2.5781, 1.5703, 2.125, 2.125, 2.0312, 1.4062, 0.8789, 3.2344, 1.3984, 2.8906, 2.7969, 2.0625, 3.0469, 2.8125, 2.3906, 1.5938, 2.0469, 3.1875, 2.8594, 2.0156, 1.8516, 2.2969, 3.0469, 2.4844, 2.2188, 3.25, 1.875, 2.375, 2.3438, 2.6875, 1.5469, 2.5, 3.1094, 2.7656, 2.5938, 2.0156, 2.9062, 2.6562, 1.0234, 2.5156, 2.7188, 3.3594, 0.7656, 2.8438, 3.3125, 0.9414, 2.2344, 3.3906, 3.5, 0.7656, 5.1875, 3.4062, 1.7812, 1.8672, 2.5938, 1.7031, 5.0938, 3.5625, 2.4219, 2.8438, 2.5312, 1.4766, 2.5156, 0.8789, 1.9219, 2.5781, 2.5, 4.6875, 2.5625, 1.5781, 1.8203, 2.9375, 2.2969, 0.8438, 0.8242, 1.7109, 2.2656, 3.25, 5.0, 2.5625, 1.6094, 2.7812, 2.7969, 2.5625, 0.8281, 1.8125, 1.1406, 2.625, 1.9688, 1.2344, 2.6094, 2.5469, 0.9258, 2.4844, 1.8672, 2.5781, 1.4453, 1.8984, 1.6797, 0.7656, 1.4609, 2.2656, 3.0312, 2.2656, 1.8203, 2.7812, 2.7188, 2.6719, 2.4844, 2.7969, 2.25, 2.9844, 2.2344, 1.3594, 2.5781, 3.0, 4.0625, 3.9219, 2.7031, 2.2812, 4.6875, 1.4609, 2.4062, 0.7969, 2.5, 2.75, 2.3906, 2.6562, 1.6797, 2.6562, 2.5, 1.9531, 5.0312, 2.2656, 2.7031, 3.0469, 3.125, 2.3906, 3.25, 2.875, 2.6719, 0.8125, 2.0, 1.7891, 2.1719, 1.9766, 1.8594, 2.3906, 3.1094, 2.6094, 2.5625, 2.9062, 3.2031, 1.6172, 1.8594, 2.8438, 1.4297, 2.5781, 2.625, 1.1641, 1.8984, 2.8281, 1.9375, 2.4531, 2.1562, 0.8047, 2.4062, 2.4844, 2.5, 3.3438, 1.7891, 1.1016, 0.9531, 4.0625, 2.0469, 3.3906, 1.4219, 0.793, 1.2969, 3.0625, 4.3438, 1.3203, 1.9297, 0.8203, 2.4375, 2.3906, 3.0156, 1.5, 1.6953, 5.6875, 2.7188, 1.5547, 1.7422, 2.7031, 2.8281, 2.4844, 2.4844, 1.7969, 2.0312, 2.9688, 2.7656, 1.3594, 2.6406, 2.3594, 1.8828, 2.4844, 1.3516, 2.8906, 2.0469, 2.2969, 1.0156, 2.4844, 0.9492, 2.4531, 1.5859, 4.9062, 1.7734, 2.3125, 1.6172, 4.8125, 1.6641, 2.6094, 2.0781, 2.5156, 1.8828, 2.5, 2.7188, 1.6953, 2.625, 1.5938, 2.125, 1.2812, 2.2188, 1.5312, 2.2344, 1.8438, 3.125, 2.375, 1.8984, 0.7773, 2.6562, 1.6953, 2.4844, 4.75, 2.9844, 2.8594, 1.4766, 5.25, 4.6875, 2.0938, 2.5, 2.1406, 2.2344, 0.7383, 2.1406, 3.1406, 2.4375, 1.5859, 2.7344, 2.0625, 2.3594, 2.3125, 2.6562, 2.5781, 1.6562, 2.4375, 1.125, 3.3438, 2.5625, 2.5312, 1.0547, 1.6562, 1.1797, 2.6094, 2.25, 1.7734, 2.375, 0.832, 2.5469, 2.3281, 2.9375, 4.75, 1.6484, 2.0156, 4.375, 2.5469, 1.2422, 2.375, 2.7969, 2.2344, 2.6719, 2.3125, 2.4688, 1.3203, 2.3906, 2.8125, 2.4531, 3.75, 1.0469, 2.5938, 2.2344, 2.25, 2.6562, 1.1328, 1.2734, 2.4375, 1.7734, 1.1328, 1.9766, 2.7031, 1.2031, 4.5312, 0.7539, 1.5391, 2.4688, 2.5781, 1.8984, 1.2734, 2.5312, 2.3125, 1.8438, 1.7109, 2.0469, 2.7812, 1.4297, 3.2656, 2.125, 1.2422, 2.4688, 1.6016, 1.2266, 1.3438, 2.7344, 2.0, 2.6094, 2.6406, 1.9375, 1.8047, 3.4844, 0.8789, 2.125, 2.7188, 2.6562, 2.2344, 2.375, 1.5234, 0.8594, 2.4375, 1.3281, 1.1953, 1.3516, 2.6094, 1.2109, 1.5469, 2.375, 2.375, 2.9531, 3.0312, 2.0625, 2.6094, 0.7852, 2.9062, 2.2344, 2.0781, 2.5469, 2.7812, 2.0625, 2.1562, 0.7773, 0.8281, 2.1406, 2.8281, 3.2344, 2.6719, 1.7422, 1.7891, 1.5625, 0.9688, 3.0, 2.375, 2.4688, 2.5156, 1.7422, 1.3594, 2.5, 2.3438, 2.6875, 4.4062, 2.0938, 2.3438, 3.9531, 1.2969, 2.0312, 2.7344, 2.5625, 0.8398, 1.8828, 1.3047, 1.1406, 2.2656, 2.6875, 1.1875, 2.1719, 1.9297, 1.8047, 2.3281, 1.7422, 2.8281, 2.2188, 2.5781, 0.75, 2.6094, 1.7188, 0.7227, 2.5, 1.5391, 2.2031, 1.5625, 0.9883, 1.1797, 1.9844, 2.0625, 1.8047, 3.0781, 1.9844, 0.7266, 2.3594, 2.4375, 2.0312, 2.5469, 2.5625, 2.4688, 2.1562, 2.5, 1.0703, 2.4531, 3.0, 0.8633, 2.2812, 2.8438, 1.7656, 1.6172, 1.8281, 2.0625, 1.9922, 2.3125, 2.7188, 4.375, 1.1719, 1.9453, 1.0, 1.625, 2.0625, 2.7188, 1.7656, 2.5781, 2.6562, 4.7188, 2.125, 2.6406, 1.625, 2.125, 2.0625, 2.2656, 2.2656, 2.75, 1.7266, 1.4375, 2.5469, 1.6016, 1.7891, 2.3281, 1.9453, 1.0469, 1.3672, 2.6875, 2.5312, 2.6094, 2.2188, 2.1875, 1.7266, 2.125, 1.6094, 2.9844, 1.5469, 2.2344, 1.9375, 2.5938, 2.4062, 1.8438, 1.6328, 2.1094, 1.5938, 1.2422, 4.1562, 1.6484, 2.2969, 2.0, 2.1719, 2.3438, 1.4219, 2.5312, 1.8203, 1.5859, 2.6719, 2.5156, 1.9688, 1.5625, 1.3828, 3.1562, 2.5625, 4.9688, 1.4375, 4.2188, 2.3594, 1.1953, 2.4844, 2.1562, 2.5312, 1.7109, 2.6719, 2.4844, 2.0781, 1.8438, 1.9531, 1.2891, 2.7656, 2.4219, 1.9844, 2.25, 2.0, 1.375, 1.9062, 2.4062, 1.8203, 1.5, 2.8438, 2.2812, 2.4688, 1.1875, 1.7734, 1.0234, 3.9844, 1.9141, 2.3438, 1.8672, 1.7891, 2.1094, 2.0781, 0.7539, 1.5625, 1.9141, 1.9062, 1.875, 3.0469, 1.8594, 0.8086, 1.2109, 1.7891, 2.5156, 1.5, 1.7109, 1.5547, 1.1875, 2.4062, 1.6641, 1.3672, 2.5312, 1.8594, 2.7656, 1.6172, 2.0, 1.6562, 0.793, 1.8672, 1.2734, 4.9062, 2.5312, 2.3906, 4.375, 2.0312, 2.0938, 1.8906, 1.6719, 2.0625, 1.0625, 1.7891, 1.2031, 2.2656, 2.7188, 1.9766, 1.3828, 2.4375, 2.7188, 2.4531, 1.7266, 2.0625, 1.9141, 0.9922, 2.6719, 1.125, 2.3594, 2.2031, 2.3906, 2.3125, 0.7695, 2.8125, 2.25, 1.4609, 1.6016, 2.2344, 2.5, 1.8984, 2.25, 2.375, 0.8867, 2.4688, 2.5156, 2.0, 1.8359, 2.7812, 0.7852, 2.0781, 2.3125, 1.7734, 0.7578, 3.2969, 1.5156, 1.0, 0.7578, 2.2344, 2.1562, 2.5469, 2.3281, 1.7578, 2.0469, 1.9297, 1.8984, 3.0, 0.9062, 2.1719, 2.4531, 2.3594, 2.5312, 2.7812, 1.8125, 1.3203, 2.3594, 2.0469, 1.5938, 1.7812, 2.3125, 1.4219, 2.2969, 1.5859, 2.375, 1.25, 2.4375, 0.918, 2.375, 1.3672, 2.4219, 2.4219, 0.8906, 1.4766, 1.2109, 2.4688, 2.4375, 2.875, 3.0, 0.8281, 1.6016, 2.0, 2.6719, 2.5781, 0.6953, 1.4609, 2.3281, 1.3906, 1.0703, 2.2969, 2.7812, 2.1562, 2.2031, 1.9375, 2.25, 2.7031, 1.2344, 0.7227, 2.4688, 1.7969, 1.9375, 1.8672, 2.3438, 4.4062, 2.2812, 3.0, 1.6953, 1.2969, 2.1094, 2.25, 1.6484, 1.4922, 1.2812, 2.4688, 2.5781, 2.1562, 2.3594, 1.125, 4.25, 1.6797, 2.1562, 1.0781, 2.3125, 2.4375, 1.5156, 0.8398, 1.9609, 2.5938, 1.7734, 1.1328, 2.5312, 1.0078, 2.1562, 1.9531, 1.4531, 1.6719, 1.2969, 5.3438, 2.0938, 1.8359, 2.4375, 1.6875, 2.5156, 0.8008, 1.5312, 2.5156, 0.668, 1.4531, 1.125, 1.8984, 2.3438, 1.7344, 1.9766, 2.9219, 1.7891, 1.6094, 2.2031, 1.4922, 2.6719, 2.4688, 1.6562, 2.2812, 1.5703, 2.6094, 1.3125, 2.5469, 2.1719, 2.0469, 1.1406, 1.9766, 2.5, 0.8594, 1.0625, 2.6094, 1.7109, 3.5312, 2.3125, 2.3281, 2.0625, 2.2031, 1.8359, 4.2188, 2.4062, 1.3359, 4.375, 2.4844, 5.0625, 2.2188, 1.0625, 1.25, 2.0312, 1.4219, 2.4688, 1.3281, 1.2109, 2.3906, 1.3672, 1.9219, 2.0625, 0.8047, 1.7344, 1.6016, 1.7734, 2.3906, 1.7344, 4.6875, 2.375, 0.8594, 4.3438, 2.0625, 1.6641, 2.8594, 2.7188, 1.4297, 2.2031, 2.2812, 2.5156, 2.4688, 2.7344, 2.0312, 1.1797, 1.9375, 1.3359, 2.3281, 1.25, 2.4531, 1.8672, 2.4844, 1.9297, 2.3438, 3.5469, 2.125, 2.1562, 1.6172, 2.0938, 1.7578, 2.2188, 2.1875, 2.1719, 2.3281, 1.6797, 2.1094, 1.9531, 2.3125, 1.6797, 2.7969, 1.6016, 2.0469, 1.6875, 0.9922, 1.6641, 0.8203, 2.4844, 1.7422, 1.7891, 2.5156, 1.2109, 1.8203, 1.9766, 1.625, 2.2969, 2.3594, 2.25, 0.793, 0.7461, 1.5078, 0.8789, 2.0469, 1.9688, 2.3594, 2.3906, 1.0547, 2.1094, 0.8242, 0.832, 1.7422, 1.7031, 2.375, 1.8438, 4.25, 1.9453, 2.4844, 0.5742, 1.8516, 1.9688, 1.3359, 2.125, 1.6562, 2.6875, 2.2344, 1.4219, 1.2188, 2.2344, 2.625, 2.375, 1.7812, 1.7266, 3.7969, 1.4062, 2.7344, 2.5, 0.6836, 1.4219, 0.7539, 1.9922, 1.6172, 2.4219, 1.8047, 0.9648, 0.8906, 1.9375, 2.1562, 1.8906, 1.2578, 1.5938, 1.9531, 2.3594, 1.3984, 1.4141, 2.4688, 0.7695, 0.8398, 2.0, 2.0625, 3.0938, 1.8984, 2.0, 1.2344, 1.4609, 1.9297, 2.1719, 2.9531, 0.8086, 1.7266, 2.3906, 1.6562, 2.7031, 0.7969, 1.7734, 2.6719, 2.0312, 1.9375, 0.9023, 0.7539, 0.8125, 2.125, 2.0156, 2.25, 1.4297, 1.4531, 1.4844, 1.8047, 1.9219, 1.4453, 1.5547, 1.8828, 1.6328, 1.7734, 1.9297, 2.1094, 0.9375, 2.4375, 1.9453, 2.2188, 1.4844, 1.3125, 1.625, 2.3906, 2.0938, 1.9688, 1.0234, 1.5, 2.125, 0.7031, 1.6172, 1.7812, 1.4141, 1.9922, 2.3438, 1.7891, 2.7031, 0.9609, 2.2969, 2.0625, 1.7266, 1.4609, 4.6875, 1.6875, 2.0, 2.4062, 2.1719, 0.7656, 2.1875, 0.5195, 0.9688, 1.6641, 0.7734, 1.1641, 2.125, 4.8438, 1.9766, 2.0469, 0.9727, 2.4844, 2.2656, 1.6484, 0.7617, 2.1562, 1.1406, 2.3438, 1.1406, 2.3594, 2.5312, 1.5938, 0.8086, 2.1094, 2.25, 1.6875, 1.5312, 1.5547, 2.0625, 1.5391, 2.5625, 1.2812, 1.1562, 1.8594, 1.2266, 1.4531, 2.6562, 1.0547, 1.0703, 1.8203, 2.1719, 2.4062, 1.8359, 1.875, 1.6797, 2.8281, 1.8672, 1.5391, 0.6445, 4.125, 1.7891, 1.1797, 2.1094, 1.25, 1.2188, 1.8203, 1.2266, 1.7031, 0.9414, 0.957, 2.5, 1.1094, 1.7109, 1.5391, 1.9922, 1.6719, 1.1172, 1.1641, 1.5859, 2.2344, 3.8281, 1.4219, 1.7109, 0.5078, 2.5781, 2.0625, 1.3359, 4.125, 0.6328, 1.7656, 2.0781, 2.1719, 2.2344, 1.9688, 1.6719, 1.6484, 1.4141, 1.1875, 1.6484, 2.2812, 1.4688, 1.9453, 2.0469, 1.7812, 0.8867, 2.25, 1.4375, 1.6797, 1.5391, 2.1719, 1.7812, 2.0469, 1.7422, 4.2188, 2.5469, 0.8242, 1.2109, 1.7188, 1.0078, 1.4141, 1.2891, 2.2656, 2.25, 1.5781, 2.5156, 2.5156, 2.2656, 1.6953, 2.2656, 2.0156, 2.0938, 3.7031, 4.0312, 1.6719, 4.5, 1.0078, 1.3594, 1.7578, 2.8125, 1.6719, 1.0781, 2.3125, 1.7656, 2.9531, 1.625, 1.5938, 1.9531, 0.7656, 1.875, 1.4297, 1.6016, 2.1719, 1.9297, 2.5938, 2.1719, 2.7344, 0.8164, 2.3906, 1.0156, 1.6094, 2.5469, 1.8984, 1.9766, 1.2266, 2.3438, 2.6562, 2.6094, 2.3281, 2.2656, 3.6875, 2.4219, 3.75, 1.1562, 1.7422, 2.2344, 0.8828, 0.8438, 0.9219, 1.5156, 1.9766, 1.8516, 1.5938, 2.2969, 2.1406, 2.7969, 2.1719, 2.375, 2.6875, 1.5703, 1.8828, 0.7656, 1.0234, 2.5312, 2.2031, 0.7344, 1.875, 2.3438, 2.25, 1.6797, 1.8047, 2.2969, 0.7656, 1.8203, 2.0938, 1.2812, 2.5312, 4.1562, 3.9062, 1.9375, 2.2344, 1.9688, 2.0, 1.4531, 1.4688, 1.5703, 2.4219, 1.9766, 1.5859, 3.5156, 2.125, 2.5938, 1.375, 1.9297, 0.7773, 2.1875, 0.7344, 1.5469, 2.3594, 1.9141, 1.4141, 2.0781, 0.7695, 1.9922, 4.5, 2.375, 2.0469, 1.5156, 1.6797, 1.8203, 1.6875, 0.7891, 2.3281, 2.5, 3.0938, 3.8125, 1.3047, 2.2188, 1.3281, 2.0469, 2.2188, 1.6172, 2.0156, 1.6953, 1.7812, 2.5, 4.7188, 1.5703, 4.8438, 2.8438, 1.1172, 1.0781, 2.375, 1.5156, 1.6562, 0.8438, 2.6719, 1.4297, 4.4062, 1.0625, 1.3359, 1.3984, 2.2812, 1.9375, 2.0625, 1.5703, 2.0, 1.5625, 1.5, 2.0781, 3.0938, 1.9844, 1.2578, 2.5312, 2.2344, 1.4844, 2.1562, 2.0, 1.5, 1.8672, 1.8359, 1.3359, 2.375, 0.8438, 1.6094, 1.3516, 1.2891, 2.2969, 0.8008, 0.793, 1.6641, 1.1406, 1.7344, 1.4844, 0.9453, 2.1562, 1.9375, 1.375, 2.0469, 1.6797, 1.2891, 1.7422, 1.9453, 0.9609, 1.8359, 1.7266, 0.8359, 2.0625, 1.5391, 2.2188, 2.0312, 1.7812, 1.75, 2.2344, 2.1719, 2.2656, 2.875, 1.1406, 2.0312, 1.5, 1.7344, 1.8281, 1.6875, 2.0156, 1.25, 1.5, 2.0625, 1.875, 0.9297, 0.9414, 2.3438, 2.75, 2.0156, 0.8477, 2.5938, 1.2656, 5.0625, 1.375, 1.1562, 1.2891, 1.7344, 1.1172, 2.4062, 2.4688, 0.9141, 1.5078, 2.25, 0.7891, 2.5781, 2.1875, 2.2656, 2.2812, 1.9531, 1.9688, 4.1562, 0.7383, 1.7969, 1.5625, 2.4375, 1.2891, 1.1016, 2.7031, 2.2188, 2.3594, 1.875, 3.5, 2.0625, 1.4375, 2.2969, 1.6562, 1.5391, 0.7539, 2.5781, 1.9531, 1.0859, 1.875, 2.5625, 2.0469, 1.7109, 2.5625, 2.0625, 1.6641, 1.625, 2.1875, 1.5859, 1.4297, 2.4375, 1.4531, 2.0156, 1.8281, 2.1719, 0.9609, 1.5625, 1.2891, 4.4375, 2.6875, 0.9688, 1.9688, 1.2969, 2.1094, 4.1875, 1.3359, 1.4766, 2.5, 2.2188, 2.6719, 2.0625, 2.0938, 1.3359, 3.1094, 4.75, 2.2031, 2.3281, 1.5, 0.8008, 1.8281, 1.8281, 0.7344, 2.0781, 1.9688, 1.9453, 1.4062, 1.25, 1.8125, 1.4688, 2.0, 0.7773, 2.0938, 0.8047, 1.0938, 1.9609, 2.1562, 1.5859, 1.8203, 3.8594, 1.2656, 2.0156, 2.5625, 2.3594, 1.8125, 0.832, 1.1719, 2.4375, 1.3438, 0.7695, 2.2656, 1.3359, 1.3906, 0.9531, 2.4688, 1.6484, 2.0469, 1.3359, 2.3438, 1.1953, 1.625, 2.0312, 2.5469, 1.9453, 2.75, 2.4844, 1.2812, 1.1641, 2.4062, 1.9688, 2.2969, 0.793, 1.4062, 1.4219, 1.1484, 1.2656, 1.9609, 1.8359, 3.8594, 1.7031, 2.5781, 1.6797, 2.0, 1.6094, 1.6562, 1.8672, 1.8438, 2.2031, 1.8438, 1.8281, 1.5781, 2.4531, 1.1641, 2.2812, 0.9844, 1.7969, 0.8047, 1.2891, 2.1562, 2.25, 1.3438, 2.4375, 1.9062, 1.8203, 1.1562, 1.3438, 4.1562, 0.9023, 2.1562, 0.7812, 2.4062, 2.0781, 1.7031, 1.5391, 1.9141, 1.2969, 1.1875, 1.0938, 2.1875, 1.3047, 1.8984, 2.4531, 2.0312, 0.6055, 1.6406, 2.4219, 1.7188, 2.1875, 1.25, 2.1875, 2.1875, 2.1562, 3.375, 0.8047, 0.7227, 2.5781, 2.2031, 1.6406, 2.2188, 0.9102, 0.7695, 2.0469, 2.1562, 2.2188, 2.0469, 1.4062, 1.7656, 1.5391, 1.6172, 2.1875, 1.2734, 2.8906, 2.125, 2.0156, 2.1406, 2.3438, 2.5312, 2.1406, 1.4531, 2.3438, 2.375, 0.7617, 1.7891, 1.7266, 0.7539, 1.9062, 1.4062, 0.793, 2.4375, 2.0625, 1.0938, 2.125, 0.7109, 2.5781, 1.9531, 0.8281, 2.0781, 1.7266, 2.4062, 1.7812, 4.125, 1.8203, 0.8125, 0.8125, 2.3594, 1.3828, 1.2812, 1.9062, 1.4922, 2.2188, 1.9531, 3.3125, 1.6875, 1.0703, 3.9531, 1.6484, 2.1094, 1.6875, 1.5703, 3.9375, 1.0391, 2.2344, 1.8047, 1.9141, 2.25, 2.4688, 1.8438, 4.125, 1.4453, 1.8125, 1.5156, 0.6758, 1.4297, 1.2891, 2.25, 2.2812, 1.0469, 1.25, 2.0312, 1.3438, 2.1719, 1.8594, 2.0156, 2.1406, 2.5, 2.25, 1.7109, 1.9453, 2.0, 2.6094, 2.1094, 1.7734, 2.4375, 1.4844, 1.5391, 1.6406, 1.5938, 1.7891, 2.2812, 2.4688, 0.5469, 2.5625, 2.1562, 1.5938, 1.4766, 1.5547, 1.8359, 1.4531, 1.9453, 2.4219, 1.6797, 2.625, 1.0078, 2.1406, 2.8125, 0.7695, 1.1094, 3.8906, 1.4688, 2.8594, 2.25, 4.0312, 1.5547, 1.2109, 1.4297, 2.1875, 0.6992, 4.375, 1.7734, 1.7656, 2.3125, 0.9922, 1.1875, 1.8828, 1.6562, 1.3359, 1.8203, 2.1719, 1.5469, 3.1094, 2.3906, 1.4531, 1.375, 1.1094, 1.0547, 0.7852, 2.2812, 0.6016, 2.2656, 1.2734, 1.7266, 0.6914, 1.7031, 1.5, 2.0938, 0.832, 2.125, 2.2344, 1.7344, 2.5312, 1.5938, 1.6953, 1.9922, 2.125, 2.1094, 2.5, 0.5391, 1.5312, 2.4844, 1.7109, 2.5, 1.3516, 4.0312, 1.6484, 2.4688, 1.5938, 2.3281, 1.6328, 0.7578, 2.3906, 0.5898, 1.4375, 1.8203, 1.1406, 1.9766, 3.7031, 1.4531, 1.6328, 1.4219, 1.3438, 0.7969, 2.1562, 1.625, 2.1094, 1.3516, 3.7969, 2.25, 2.0, 1.9219, 0.7344, 2.2812, 1.0703, 2.0781, 1.9062, 1.3672, 1.7031, 2.1719, 1.6094, 1.4609, 2.0625, 1.1172, 0.8008, 1.8438, 2.2031, 2.0469, 1.2266, 2.0, 2.3125, 1.5859, 1.6875, 0.6289, 1.0469, 1.3125, 1.2422, 1.4062, 1.1406, 1.6328, 2.0469, 1.1328, 1.8125, 2.7031, 1.4688, 2.2031, 0.6445, 2.2344, 2.2344, 1.9609, 1.0547, 3.9531, 2.0938, 1.4766, 1.2344, 2.0625, 1.7891, 1.2188, 2.0625, 1.8359, 1.625, 1.7188, 2.1719, 1.3828, 1.5312, 1.6484, 0.75, 2.375, 2.3125, 2.1875, 1.6562, 2.375, 0.8047, 2.4531, 1.4297, 1.4219, 2.8594, 2.6406, 1.1406, 2.125, 0.8086, 1.7266, 0.8086, 1.7734, 1.5938, 1.3516, 1.6797, 1.0625, 0.9883, 1.5625, 1.125, 2.0625, 2.6094, 1.7266, 1.3281, 1.6719, 1.9219, 2.375, 1.0234, 2.0, 0.9766, 2.2344, 1.5312, 0.9219, 2.0312, 1.9609, 2.0469, 1.8594, 1.875, 0.8438, 1.2422, 1.9766, 1.4766, 1.3906, 1.875, 1.6641, 2.4375, 1.2422, 1.4141, 2.125, 1.4062, 0.9844, 3.0938, 1.5469, 1.3125, 1.7656, 2.2812, 1.7031, 1.8203, 1.4219, 1.3828, 0.8711, 2.1094, 2.6562, 0.9492, 2.1094, 2.2656, 1.8594, 1.7344, 2.0, 1.2812, 1.6719, 2.0312, 0.9609, 1.6719, 4.5938, 0.793, 1.7578, 2.1562, 1.3125, 1.4609, 4.0625, 1.6016, 0.8672, 1.8359, 1.4844, 2.5, 0.9414, 2.5781, 1.6641, 1.4688, 2.0938, 1.6875, 1.8203, 2.5938, 1.9062, 2.25, 1.6016, 1.3672, 1.2109, 1.6094, 1.5078, 1.4844, 0.5273, 3.8281, 4.2188, 2.4688, 1.5391, 3.8125, 4.625, 1.6328, 2.3906, 2.2969, 0.8867, 1.9062, 1.5703, 1.4531, 1.3047, 2.4844, 2.2656, 1.7812, 1.8906, 4.0, 1.7031, 1.5625, 1.3828, 1.9219, 1.4531, 2.7031, 1.2422, 1.9453, 2.0781, 1.3359, 1.1797, 1.4297, 1.9688, 2.9219, 2.4062, 1.6641, 2.4219, 1.7109, 1.6328, 1.5078, 0.7578, 2.2344, 1.4531, 4.375, 2.3281, 1.8672, 1.7188, 1.9844, 1.75, 0.9414, 1.6875, 1.2656, 1.625, 2.1094, 2.0156, 1.6172, 1.3594, 1.6328, 1.5156, 2.4531, 1.5, 3.0312, 2.1406, 0.8633, 1.2656, 1.2734, 2.4219, 2.2812, 2.2344, 4.25, 0.7188, 1.3594, 3.9219, 1.5078, 1.3516, 2.1406, 2.6562, 1.0938, 1.1484, 2.125, 1.1797, 2.0312, 1.2891, 0.7734, 2.3125, 1.7656, 1.4141, 1.2734, 1.1484, 1.5078, 1.8516, 1.7109, 1.1328, 1.3828, 0.7031, 2.375, 1.7031, 2.1406, 3.8438, 2.6094, 1.7812, 1.5625, 3.9062, 1.6406, 1.2109, 1.7656, 2.1094, 2.2969, 2.2812, 0.8047, 2.0938, 2.0469, 1.6641, 2.25, 1.6172, 1.9688, 2.0312, 1.3047, 2.3594, 1.8125, 2.0, 2.2969, 0.8359, 2.25, 2.4062, 2.1094, 0.8555, 0.7734, 1.5391, 1.9844, 2.2031, 1.9219, 1.7969, 1.7422, 3.0, 2.2031, 2.3281, 1.3203, 1.2344, 0.7148, 1.9453, 1.2656, 2.375, 3.5781, 2.1719, 2.1875, 1.4141, 2.5, 2.125, 0.7539, 1.6719, 2.2969, 2.2344, 1.6719, 2.3281, 2.1719, 2.4375, 1.4375, 1.6719, 2.1094, 1.4219, 2.125, 1.6172, 2.2344, 3.1719, 2.375, 1.9453, 2.3906, 2.7656, 1.5, 1.9141, 1.4688, 2.25, 2.2969, 1.6016, 2.1875, 2.3281, 1.8359, 2.25, 1.4688, 1.1875, 2.4375, 2.0, 1.8125, 1.5078, 2.3594, 4.6562, 2.0156, 0.9648, 2.1875, 1.8125, 1.6641, 1.2812, 0.9102, 1.7578, 1.75, 4.625, 1.6016, 0.9492, 3.7969, 3.875, 1.4766, 1.7109, 2.0, 2.4688, 1.6641, 1.4766, 1.1953, 1.9141, 1.4453, 2.2344, 1.9141, 2.25, 0.9844, 2.4531, 1.3516, 2.2656, 2.1094, 0.8438, 4.1875, 2.4531, 2.5, 4.0, 0.9141, 1.8984, 2.1406, 1.9375, 2.125, 3.9219, 1.5547, 1.6328, 1.7188, 0.7344, 1.5781, 2.2031, 2.4375, 1.8203, 2.4375, 2.7812, 1.6562, 1.8594, 4.4375, 2.375, 2.0938, 1.6328, 1.9766, 2.3594, 0.7148, 1.1172, 1.7422, 1.3906, 1.4375, 2.0312, 1.3047, 1.0781, 2.25, 0.7852, 1.6797, 1.7656, 1.7109, 0.5469, 1.2656, 1.2656, 1.1484, 2.3125, 0.8125, 1.625, 1.7266, 1.5859, 2.1094, 1.7656, 0.7656, 1.3359, 2.2812, 1.5391, 1.8125, 1.3359, 2.2188, 1.3828, 1.2969, 1.5703, 1.5078, 0.9883, 2.25, 1.1406, 2.25, 2.4219, 0.918, 0.8945, 2.1875, 2.125, 1.5, 1.375, 2.0781, 0.6719, 1.5938, 2.0156, 1.0938, 2.0, 1.7891, 1.5547, 0.75, 1.9062, 1.8516, 2.5156, 2.5469, 1.5156, 1.2969, 0.9297, 1.7891, 1.1562, 0.918, 1.8438, 1.1484, 0.7188, 1.8203, 2.625, 1.5234, 1.7109, 2.0938, 0.75, 1.6875, 1.4922, 1.5312, 0.8594, 1.5703, 2.3906, 1.75, 2.1719, 1.8672, 1.5547, 2.625, 1.5625, 1.625, 3.7812, 1.8594, 3.2656, 1.8359, 2.4531, 2.4062, 2.2656, 1.8203, 1.9375, 1.5, 3.2031, 0.5859, 1.3203, 2.3281, 1.6484, 0.7773, 2.2812, 1.6328, 1.9375, 1.5703, 1.2031, 2.3594, 1.2422, 0.9883, 2.2031, 0.9492, 2.0625, 2.3125, 1.7266, 2.2656, 4.3438, 0.8555, 1.9844, 1.6484, 1.6016, 1.1875, 3.9531, 3.9219, 0.7383, 1.7031, 2.0, 2.4062, 1.7578, 2.6875, 0.8125, 0.7852, 1.6094, 1.7969, 1.8125, 1.9766, 2.1094, 1.5, 1.7266, 1.5547, 2.2344, 1.5781, 1.6016, 1.7734, 2.2969, 2.2188, 2.125, 2.0625, 3.9219, 1.7656, 2.1406, 1.5781, 1.1484, 1.1094, 2.0781, 1.9141, 2.1719, 2.375, 0.7812, 2.1406, 0.75, 1.3906, 1.625, 1.25, 2.5156, 0.8203, 2.1562, 0.8203, 1.0312, 1.375, 2.0, 0.9062, 2.1562, 2.3594, 1.3906, 2.125, 1.8672, 2.4844, 2.5156, 1.8359, 0.9961, 1.6016, 1.8125, 1.2734, 2.4375, 4.2812, 2.6094, 1.2188, 2.1875, 1.8984, 2.0312, 1.3359, 4.25, 2.0156, 1.7656, 1.9766, 1.1328, 1.4922, 1.4453, 2.2031, 1.7266, 1.6328, 2.0312, 1.1641, 2.1719, 1.3828, 0.8477, 1.8594, 1.3672, 0.8477, 2.1875, 1.1641, 1.7109, 1.6406, 1.9922, 1.6953, 0.7812, 1.5859, 2.6562, 0.9688, 1.4375, 1.5234, 1.0078, 2.375, 1.6406, 1.9531, 2.2656, 0.8906, 1.8047, 1.9219, 2.2812, 1.1719, 1.6562, 0.7227, 1.8125, 1.3438, 1.5703, 1.9297, 0.9023, 1.7812, 0.8203, 2.7656, 0.832, 2.2344, 2.0781, 1.4766, 2.2969, 1.2734, 1.7812, 5.1562, 2.5, 2.5625, 1.3359, 0.8828, 0.9609, 2.1719, 2.3125, 1.125, 1.6016, 1.8672, 1.9062, 1.9062, 1.4609, 2.5312, 1.625, 0.9492, 1.8594, 1.4062, 0.8867, 1.1484, 1.0156, 2.25, 1.9844, 0.7812, 1.4062, 1.1094, 1.3828, 1.6719, 1.7812, 1.6094, 1.7734, 1.3281, 0.8242, 1.4141, 2.2031, 0.9219, 1.4375, 1.6172, 1.1797, 2.2031, 2.3281, 0.5938, 0.7539, 2.875, 2.4062, 1.1094, 1.7734, 1.0625, 1.6406, 0.9531, 2.1562, 2.3438, 2.2031, 0.8477, 0.8633, 1.6719, 2.2031, 1.4766, 1.8828, 1.7188, 1.6641, 0.918, 1.75, 1.2891, 1.6406, 1.6406, 1.3984, 2.4844, 1.5078, 1.7656, 0.8242, 1.7812, 2.375, 1.0078, 1.8438, 2.4375, 1.3672, 1.0625, 0.793, 1.3281, 1.8125, 1.2891, 2.6094, 1.2812, 0.7891, 1.2422, 2.1719, 1.2891, 1.5078, 3.9375, 1.3672, 4.1562, 1.9219, 1.4375, 1.1406, 2.0312, 1.0078, 1.5, 1.8359, 2.3125, 2.4844, 1.4453, 0.9922, 1.5, 3.625, 1.2578, 2.25, 2.3906, 1.6094, 1.0234, 2.2969, 1.9062, 2.2812, 1.5, 1.2891, 1.5312, 1.3203, 1.9922, 2.3438, 1.3594, 2.4375, 2.4531, 2.25, 2.3281, 2.0, 2.4375, 1.1719, 2.2344, 1.5078, 2.2656, 1.75, 0.793, 2.2969, 4.1562, 1.875, 1.8828, 1.4766, 1.7344, 2.1719, 2.5, 3.5, 1.6562, 2.5625, 0.8438, 0.8477, 2.3281, 1.1484, 2.3125, 2.5, 1.4219, 2.8594, 2.0156, 2.4688, 2.2812, 0.793, 1.0469, 2.4531, 1.5, 0.9531, 0.9141, 0.7617, 1.3594, 1.1641, 0.9023, 1.4922, 1.0234, 1.1797, 0.7383, 1.5781, 0.9727, 2.0, 0.75, 2.0781, 1.6406, 1.8906, 0.7578, 1.5469, 1.3281, 2.0312, 2.3438, 1.3594, 1.5781, 2.2656, 1.8438, 2.0938, 1.4609, 1.6562, 2.6094, 1.7656, 1.1797, 2.0312, 1.9844, 1.9062, 2.0781, 2.3281, 2.2656, 1.0312, 1.0234, 1.0234, 2.1094, 1.3906, 2.6406, 2.0156, 1.9609, 1.3906, 0.875, 1.7109, 1.5625, 2.2656, 1.2812, 1.8906, 0.8398, 1.3203, 1.8516, 2.5938, 1.6328, 1.5078, 1.0781, 1.6797, 1.7812, 2.0781, 1.7656, 1.3359, 1.2344, 0.6484, 2.1094, 2.5938, 1.0938, 2.25, 1.8125, 1.4609, 1.1016, 3.6719, 2.9844, 2.2812, 1.2266, 2.0312, 1.875, 1.3984, 0.9023, 2.2031, 0.9102, 2.2812, 2.5156, 1.4844, 2.2031, 2.4375, 1.8828, 1.9609, 1.7266, 1.3125, 1.4062, 0.5312, 0.9336, 0.832, 1.9609, 1.1094, 1.3906, 1.375, 1.4922, 0.7812, 0.8672, 2.4844, 1.1172, 1.4453, 2.0938, 0.8477, 1.3438, 1.125, 2.1719, 1.6016, 1.2812, 2.4688, 3.625, 1.4453, 3.6562, 1.7578, 2.4375, 1.8125, 2.125, 2.0938, 1.6172, 1.0703, 2.8125, 2.2031, 3.7344, 1.9609, 1.6797, 1.4297, 1.2891, 0.9727, 1.7109, 4.4375, 2.4844, 0.9531, 1.4297, 1.8594, 1.3281, 3.9844, 3.3281, 1.7031, 2.3281, 2.2656, 2.0156, 1.1172, 2.0156, 1.4688, 4.1875, 1.1875, 1.9062, 2.0625, 1.8359, 0.5312, 1.8516, 4.0, 1.5469, 1.25, 2.25, 1.2969, 1.9844, 0.8789, 3.7969, 2.0938, 2.0469, 1.1875, 1.2734, 3.7188, 1.3125, 1.0234, 1.2578, 2.3438, 1.6172, 4.4375, 1.7422, 1.1094, 1.0391, 1.0625, 1.625, 1.7422, 1.2266, 1.7422, 1.9844, 1.5625, 1.3125, 1.7578, 1.5078, 2.5469, 1.8281, 1.9609, 1.4609, 1.5938, 1.5625, 1.375, 1.3203, 2.4844, 1.25, 1.375, 2.0469, 2.0, 1.4453, 4.8125, 2.0469, 1.4375, 2.3125, 2.4531, 3.0312, 1.4844, 2.375, 0.875, 1.2656, 1.0078, 4.5312, 4.75, 1.9688, 1.0391, 4.125, 1.9766, 0.8047, 1.6953, 2.1719, 1.4922, 0.8789, 1.2266, 2.125, 1.5, 0.8242, 1.7969, 4.5, 4.0, 0.9219, 2.0156, 2.3906, 1.2031, 1.9297, 1.7344, 2.2656, 1.7344, 2.0938, 0.8242, 1.7578, 2.2656, 2.2031, 0.9453, 1.2656, 2.5312, 1.7188, 0.8477, 0.9922, 1.2812, 2.3125, 0.75, 1.1328, 1.125, 1.9062, 1.5781, 2.3125, 0.9492, 1.4844, 0.9102, 2.375, 2.1406, 1.8047, 1.6562, 0.9062, 2.2344, 2.3281, 1.0938, 1.4922, 4.2188, 2.1406, 2.1875, 1.3359, 2.1875, 1.3438, 1.2344, 2.3125, 2.3594, 1.3516, 1.8516, 1.875, 1.6406, 1.5625, 1.5234, 1.1641, 2.2031, 1.4688, 2.3594, 1.3125, 1.1641, 2.0781, 1.5938, 1.5391, 1.7266, 1.5156, 1.0469, 1.3672, 4.125, 2.2656, 2.3438, 2.0156, 1.5859, 2.3906, 0.8516, 1.5156, 1.5156, 1.8125, 1.9688, 1.7266, 1.2344, 1.9844, 1.7734, 2.1094, 1.1719, 2.0625, 2.1719, 1.8281, 1.7812, 1.8359, 1.3281, 1.6484, 0.8008, 1.6094, 1.7969, 1.9844, 1.9766, 1.7344, 2.0469, 1.7891, 1.25, 2.3125, 2.3281, 1.625, 2.1719, 0.7617, 1.5156, 2.2188, 1.6328, 1.0859, 2.4219, 1.1406, 2.1562, 1.2812, 2.1875, 4.4688, 1.6406, 1.8984, 1.7891, 1.5312, 1.4297, 2.0781, 1.4531, 1.125, 0.8242, 1.9297, 0.9766, 1.6562, 1.5625, 0.7734, 1.5781, 1.5938, 2.1094, 0.7891, 1.5469, 1.7344, 0.9062, 1.875, 3.9531, 1.6016, 1.7422, 2.1562, 2.1719, 2.6875, 2.625, 1.3438, 0.7578, 1.4766, 1.4062, 1.875, 3.8594, 2.3125, 2.0156, 1.7734, 2.4688, 1.0859, 2.0625, 1.5547, 1.1094, 2.375, 1.6016, 2.1406, 2.6406, 1.2891, 1.6094, 1.8359, 0.8633, 2.3594, 1.2812, 2.8906, 1.6641, 1.8672, 1.8125, 0.9844, 1.5547, 2.0625, 2.125, 1.1094, 1.2812, 1.3281, 1.2031, 1.6016, 0.7969, 1.6562, 1.2969, 0.9375, 1.2578, 2.3125, 1.7031, 1.4375, 1.9766, 1.9141, 4.0, 1.7656, 1.5234, 1.3906, 1.375, 2.2344, 2.5156, 2.25, 0.9961, 1.6328, 1.1484, 1.9453, 2.1094, 1.4766, 1.2188, 2.3438, 1.7031, 2.4219, 1.2578, 0.7305, 1.4297, 1.3984, 0.9609, 1.2422, 2.0781, 1.5703, 1.7578, 1.5547, 0.9102, 1.7656, 2.1719, 1.1719, 0.9258, 1.625, 1.7422, 1.7109, 1.4375, 1.7188, 2.2812, 0.9023, 1.0078, 1.7109, 2.0625, 0.8398, 2.0781, 2.5781, 1.4609, 2.2656, 1.7031, 1.7656, 2.2031, 1.8438, 2.25, 2.4219, 1.2188, 2.375, 3.8281, 1.5078, 2.0, 1.2578, 0.7461, 2.0156, 1.4219, 1.3359, 2.3438, 1.625, 1.6172, 1.9609, 2.2812, 2.0469, 1.4375, 0.8398, 1.6172, 1.9688, 1.3828, 4.875, 0.7383, 2.0, 1.9844, 2.5312, 3.2188, 0.7852, 2.6562, 1.5547, 1.7266, 1.4922, 3.5469, 0.875, 0.957, 2.5469, 2.9688, 1.6953, 2.0781, 1.4844, 1.6953, 1.6406, 1.3125, 1.7734, 1.25, 2.4375, 1.2656, 1.875, 1.9609, 1.3828, 2.4688, 2.0938, 1.4219, 1.9531, 1.2422, 1.625, 1.2891, 0.9375, 1.3281, 2.8125, 1.9844, 1.3203, 1.7344, 4.75, 0.9258, 2.125, 1.8125, 1.1328, 2.8281, 1.3281, 2.4375, 1.9375, 1.3906, 1.8125, 2.4844, 1.7344, 1.7422, 2.0938, 1.4766, 0.9414, 1.0625, 1.9844, 1.75, 0.7266, 2.125, 2.2812, 1.2031, 2.4062, 1.9609, 1.7266, 1.7656, 1.4375, 0.8008, 0.8477, 1.6406, 0.7344, 2.6562, 1.0547, 2.3125, 2.4062, 1.8438, 1.25, 1.7266, 1.6641, 1.8359, 2.3281, 2.4375, 2.2812, 1.3047, 1.8438, 1.4297, 1.875, 2.2812, 1.1797, 2.3125, 1.0078, 1.1094, 1.7422, 1.9766, 0.5312, 2.2031, 1.3594, 1.6172, 1.3281, 2.4219, 1.7891, 1.9688, 0.8945, 1.9609, 1.1875, 1.8203, 1.4297, 2.2812, 1.5234, 2.3125, 2.2969, 1.7109, 2.0, 1.1328, 1.3203, 1.7188, 2.4375, 1.5, 1.5078, 2.5312, 2.6406, 1.0859, 0.5742, 2.125, 1.0156, 1.4922, 1.7578, 0.7227, 1.5469, 2.1875, 4.8125, 1.3438, 1.4297, 2.3281, 0.8945, 1.0312, 2.3438, 1.1719, 2.1406, 0.9531, 1.7031, 1.3047, 1.9766, 1.6797, 1.3125, 1.1719, 1.2422, 2.3281, 1.6641, 1.7969, 1.5859, 2.5781, 1.8203, 0.9805, 1.4766, 1.3359, 2.3594, 1.125, 1.9609, 2.3125, 2.3906, 2.0156, 0.9883, 1.3828, 3.0781, 2.25, 1.7266, 1.4844, 4.0938, 2.125, 1.2578, 1.8125, 2.375, 2.2031, 3.1875, 1.7969, 0.7773, 1.3672, 0.8633, 0.9141, 1.5078, 1.875, 2.2031, 1.0938, 1.2812, 1.7109, 1.8906, 1.7031, 1.1797, 1.3359, 1.2656, 2.0781, 5.0312, 0.9648, 2.3125, 1.9375, 1.7734, 1.8125, 1.9609, 1.3828, 1.4141, 1.3359, 1.8672, 1.8672, 2.3906, 1.625, 1.7891, 1.8281, 1.8281, 1.8203, 1.5781, 1.4844, 0.8398, 1.2734, 1.8281, 2.0938, 1.8047, 2.2969, 1.3594, 2.5156, 1.3906, 2.5, 1.0469, 1.2344, 1.5078, 3.5625, 1.5, 3.875, 0.4785, 1.5312, 0.7812, 0.7539, 1.2734, 0.7344, 1.6797, 2.2969, 2.1562, 1.375, 1.6406, 1.7266, 1.7188, 1.5156, 1.5234, 1.7969, 2.375, 1.6719, 1.5391, 0.7656, 0.707, 1.1328, 1.3438, 2.1562, 1.8203, 1.4922, 1.3359, 1.5703, 0.9297, 2.1719, 2.1875, 2.25, 1.1719, 0.6719, 1.7422, 1.75, 2.2344, 1.5703, 2.2656, 1.6094, 2.0156, 1.5312, 1.7266, 2.3125, 2.375, 2.6719, 2.3906, 1.1953, 1.8203, 0.7383, 1.7422, 0.793, 1.3438, 1.6172, 1.5547, 4.3438, 1.2266, 1.9375, 3.5938, 1.5625, 1.8516, 1.3281, 0.7188, 1.6406, 2.2812, 0.8633, 2.1094, 0.9531, 1.625, 3.4844, 4.0938, 2.2031, 2.2969, 1.3125, 2.0469, 1.3516, 2.1875, 1.8125, 2.3281, 2.2344, 1.9609, 1.3203, 2.4219, 2.3594, 0.6328, 1.2578, 1.8203, 1.8125, 1.9531, 1.9922, 1.8828, 1.6641, 1.4844, 1.8281, 1.3047, 1.0938, 1.6562, 1.0391, 1.875, 0.8438, 1.9297, 3.2031, 2.125, 2.5938, 1.3906, 1.3672, 1.9375, 1.7188, 1.2891, 0.7188, 2.2031, 2.25, 1.375, 0.9102, 1.3828, 2.7188, 2.1406, 1.6328, 2.5469, 2.9219, 1.3359, 1.4141, 4.4062, 1.375, 0.8359, 4.0, 1.5781, 2.2812, 1.0078, 0.75, 1.0547, 1.8984, 1.3906, 1.2891, 2.6094, 2.4219, 0.7227, 1.4844, 1.5469, 1.6719, 1.5078, 3.1406, 0.8594, 1.7188, 2.0, 2.4844, 1.4375, 1.8125, 1.7422, 1.2266, 2.2188, 2.0156, 0.7812, 1.4219, 2.1094, 1.0469, 2.375, 1.5781, 1.3047, 0.9336, 1.2109, 2.2031, 2.0469, 1.6562, 3.875, 1.9766, 1.0391, 1.625, 1.2109, 1.9922, 2.4375, 1.8203, 1.2031, 1.1641, 3.75, 2.3594, 1.8594, 0.6992, 1.4766, 1.8203, 3.4062, 1.6641, 2.4531, 1.7891, 0.8438, 1.625, 1.375, 0.9922, 2.0, 0.7578, 0.7617, 2.4375, 1.7578, 2.1094, 3.8281, 0.8438, 2.0469, 1.3516, 0.8281, 1.6016, 1.4141, 1.3516, 1.8438, 1.4844, 0.8633, 1.8828, 2.1406, 2.4062, 1.3203, 0.7969, 1.6562, 1.4531, 1.1953, 1.4922, 0.9609, 2.2344, 2.0625, 2.375, 2.1562, 1.6172, 1.5078, 2.2031, 1.375, 1.9609, 1.9922, 2.2188, 1.9531, 1.5234, 1.5156, 2.0312, 1.3672, 4.5, 2.7031, 1.3438, 2.0938, 4.0312, 1.7266, 1.3594, 2.3906, 1.8281, 2.7344, 1.2188, 1.8047, 0.8984, 2.1094, 1.9531, 2.5469, 2.3125, 1.5859, 0.8438, 2.1406, 1.5703, 1.5625, 1.5547, 2.0469, 0.7656, 2.6719, 1.7734, 2.2188, 1.7891, 2.0781, 1.1484, 0.9336, 1.7109, 1.8906, 2.4688, 1.3516, 1.4375, 1.0703, 1.6406, 2.2188, 3.0, 1.6641, 1.7656, 2.6406, 1.8672, 1.9219, 1.5938, 1.9219, 1.4219, 0.6953, 1.8125, 4.0312, 1.5469, 2.875, 2.1875, 2.2969, 1.9609, 1.7812, 1.2969, 1.0938, 3.7188, 1.1094, 2.2812, 1.1406, 0.8281, 2.2656, 1.2656, 1.4453, 1.8047, 0.7266, 0.8789, 1.2812, 2.2812, 1.5234, 1.5859, 1.9844, 1.4922, 2.6719, 1.8906, 1.7344, 2.0781, 2.2031, 1.9531, 3.5625, 1.0391, 1.1016, 0.9844, 2.2812, 0.793, 1.6797, 2.1719, 2.4375, 1.7969, 1.75, 1.3203, 1.7734, 2.0938, 1.6328, 1.6797, 1.5, 2.1406, 2.0469, 1.3438, 2.5781, 0.8438, 1.2188, 1.5156, 2.2188, 1.8672, 1.4219, 1.3516, 1.4141, 1.2109, 1.1328, 2.0156, 1.7812, 2.0625, 0.8203, 1.0781, 1.0391, 1.8359, 2.1562, 1.9844, 1.1562, 2.2969, 1.7031, 1.9375, 2.1406, 0.7617, 1.4766, 2.5156, 1.6484, 1.7656, 1.2969, 2.0625, 1.6406, 2.5781, 1.5938, 1.4531, 1.7969, 2.4375, 3.8906, 2.125, 0.7305, 1.4844, 1.9141, 1.3438, 1.8047, 0.9766, 1.7344, 2.2656, 2.1094, 1.8672, 1.3203, 2.1094, 1.5859, 2.125, 1.2891, 1.7969, 1.625, 1.4297, 1.5859, 2.1406, 1.5469, 1.625, 1.6562, 1.8203, 1.5625, 1.3516, 1.5469, 2.4375, 1.5234, 2.375, 0.7773, 1.9688, 4.25, 1.6328, 0.7617, 1.6016, 2.2812, 2.4219, 2.3125, 1.7969, 2.5625, 1.4141, 1.3203, 4.5, 1.4141, 1.8281, 4.0938, 0.6758, 1.0625, 2.5312, 1.5, 0.832, 2.1875, 0.8086, 2.3906, 1.1172, 1.6016, 1.4688, 1.4219, 1.0469, 0.8281, 4.4375, 2.5, 0.832, 2.1406, 0.9141, 1.6797, 2.1719, 2.0312, 2.25, 1.9844, 2.7188, 1.25, 1.5781, 1.6328, 2.1719, 1.3359, 2.3281, 1.7266, 0.7695, 0.957, 1.5547, 1.8672, 1.0547, 2.2344, 2.625, 2.3125, 1.0391, 1.1484, 2.5781, 1.5391, 1.3203, 1.9219, 1.3438, 1.3594, 1.2188, 1.3281, 1.9141, 2.1406, 1.3359, 0.8789, 1.2812, 1.1094, 1.4609, 1.2891, 1.4141, 2.0469, 2.1094, 1.0156, 1.0469, 2.4062, 1.6172, 1.8984, 3.2031, 1.8516, 1.4375, 1.4844, 0.9258, 1.9375, 1.5781, 0.7773, 1.5547, 2.5312, 2.1875, 2.0312, 1.75, 1.5078, 1.3906, 1.6406, 1.4297, 2.0312, 1.7578, 2.0156, 1.8906, 1.8984, 1.4922, 1.7969, 1.6953, 2.3438, 1.5547, 1.3203, 0.9336, 1.4766, 1.3828, 1.2969, 2.3281, 1.4062, 1.1094, 2.4219, 1.9609, 0.75, 1.9219, 2.1406, 1.4297, 3.7969, 1.4688, 2.2969, 2.0312, 1.4297, 1.5781, 2.2969, 2.0625, 1.7109, 1.1172, 0.9141, 0.875, 1.5625, 1.6484, 1.7266, 2.0312, 1.8438, 0.7812, 1.5391, 1.8984, 1.3359, 1.0703, 1.3047, 1.4297, 2.1562, 2.5156, 1.3906, 1.7344, 2.2031, 2.0625, 2.6875, 0.5859, 0.7891, 1.5625, 1.3906, 1.6719, 2.0781, 1.1328, 1.4219, 1.5625, 1.2031, 1.2344, 0.7852, 1.9453, 1.2891, 2.3438, 1.9062, 1.5859, 2.8594, 1.5078, 1.6562, 1.1797, 1.7031, 0.8203, 1.3594, 1.8672, 1.4688, 0.7773, 0.7422, 1.7031, 1.8438, 2.1406, 2.1875, 1.3047, 0.9336, 4.5625, 2.3125, 3.1875, 1.3359, 1.75, 1.5547, 1.0859, 0.7578, 2.125, 1.5078, 2.0312, 1.6719, 1.6953, 2.375, 0.9688, 0.9062, 1.5859, 1.4219, 2.4375, 1.8047, 2.75, 2.0312, 2.4688, 1.7891, 2.2969, 0.9688, 2.3438, 1.4453, 0.8555, 2.2656, 1.7578, 2.0156, 1.7422, 1.8672, 2.125, 3.4531, 1.5391, 1.9531, 2.1094, 1.2656, 2.4062, 1.9453, 2.9219, 2.5938, 2.2344, 1.3359, 2.3125, 1.5156, 1.0703, 1.3828, 2.3125, 0.6641, 2.0625, 4.5625, 1.4766, 3.5781, 1.9453, 1.3281, 1.5781, 1.6172, 0.8477, 1.8438, 1.5234, 1.7344, 2.375, 2.2344, 1.1641, 1.3203, 1.7344, 1.5391, 1.7812, 1.4766, 1.5, 1.9766, 1.25, 2.2344, 1.375, 0.8086, 1.5078, 1.6484, 1.0078, 1.3516, 2.0781, 1.8594, 1.8906, 0.7305, 1.5703, 1.8047, 1.3047, 1.3906, 2.125, 1.7109, 4.3438, 0.7656, 1.9375, 0.8086, 1.5625, 1.4453, 4.4688, 1.1016, 2.0156, 2.3594, 0.7812, 0.7539, 2.2656, 2.125, 1.7891, 1.5, 1.2422, 1.6953, 2.3281, 1.6875, 1.5859, 2.1562, 1.2891, 1.9766, 1.4922, 4.5, 1.9766, 2.0, 2.0156, 3.75, 1.3516, 2.3906, 3.0, 2.125, 2.625, 0.7812, 2.0312, 1.6641, 2.8281, 0.8984, 1.1719, 0.7305, 1.7969, 4.625, 1.4141, 1.3906, 1.7422, 1.4688, 0.5703, 1.3516, 2.2656, 2.4688, 1.1172, 1.5469, 2.4531, 1.0391, 2.0938, 1.2266, 1.7031, 0.8242, 1.8203, 4.5, 1.7188, 1.0078, 4.9688, 2.0469, 0.7539, 2.2656, 2.0312, 2.0312, 1.9922, 2.3281, 2.2031, 0.7969, 1.7734, 1.6328, 1.2344, 1.6719, 2.1562, 2.2344, 4.375, 0.8516, 2.2031, 0.7891, 1.1172, 3.125, 1.2734, 1.3438, 0.9688, 1.8906, 1.5703, 1.7031, 1.6562, 2.0469, 2.0625, 1.7734, 0.8477, 1.4375, 1.3906, 0.8203, 2.0469, 0.8242, 2.0312, 2.0156, 3.6875, 2.0, 2.0938, 2.1406, 1.1562, 1.5156, 2.6094, 2.5938, 0.9453, 1.9844, 1.9141, 0.8359, 1.5938, 1.7891, 2.1094, 1.0469, 2.0938, 2.2344, 1.8125, 0.8281, 1.375, 1.7031, 1.1641, 1.2422, 1.1016, 2.0469, 2.1562, 2.0781, 1.6797, 0.8555, 1.6719, 0.8086, 1.4375, 1.3516, 1.5312, 1.7734, 0.793, 0.8672, 0.7891, 1.4375, 2.4062, 2.0938, 2.1094, 1.9062, 0.6562, 1.4062, 1.9922, 1.8438, 2.3281, 0.7031, 1.1016, 3.5938, 1.6094, 1.6406, 2.3125, 2.2812, 3.6719, 1.6094, 2.1562, 1.5703, 2.2188, 0.8789, 1.2734, 2.2812, 2.125, 2.9219, 1.8047, 0.832, 0.8164, 2.25, 1.8203, 1.1016, 1.0938, 1.8672, 1.3516, 1.4844, 0.7773, 2.2344, 2.25, 1.3594, 1.8594, 1.4766, 2.125, 1.9609, 2.6094, 1.8906, 1.3203, 1.25, 2.1562, 1.9609, 1.3828, 2.1406, 1.5, 1.0312, 1.7969, 1.6094, 1.3047, 0.7539, 1.7188, 2.1719, 1.5859, 1.6094, 2.5156, 1.0156, 1.2188, 1.0312, 1.4219, 1.4922, 2.0781, 1.1797, 1.4844, 0.4844, 1.75, 1.5156, 2.1406, 0.8438, 1.5703, 1.8516, 2.1719, 2.3281, 1.2188, 1.5, 4.5625, 2.3438, 1.3438, 1.1016, 1.4844, 1.4922, 2.0469, 1.8125, 1.9609, 2.2969, 0.8086, 4.875, 0.8672, 2.7031, 1.4688, 1.9531, 4.5312, 1.875, 1.2578, 2.0, 0.8867, 2.2188, 1.4531, 0.5469, 1.6016, 1.3359, 1.6641, 1.5234, 0.8047, 2.9062, 0.75, 2.7656, 1.9844, 0.6719, 2.6406, 2.4531, 1.3359, 2.3906, 1.9219, 2.0625, 1.0156, 1.5547, 1.5859, 1.5, 2.25, 2.5781, 1.1641, 1.7109, 2.4688, 1.25, 0.6602, 1.9609, 2.5625, 2.1719, 2.4375, 0.7656, 1.8281, 0.7188, 1.9609, 1.1641, 1.3516, 1.5781, 2.2344, 2.375, 1.5469, 0.8555, 0.707, 1.7656, 1.8438, 1.0078, 2.0, 1.8984, 1.5156, 2.8125, 1.4297, 2.9688, 2.1406, 2.1875, 3.6562, 3.1875, 1.5312, 2.1875, 2.1719, 1.2188, 2.2344, 4.0312, 1.3359, 1.3281, 1.2891, 2.2188, 1.9375, 1.5469, 2.0625, 2.4062, 1.8594, 1.2656, 2.2188, 2.6094, 1.6406, 2.5312, 1.5078, 1.3438, 0.7969, 1.1094, 1.1484, 1.3516, 1.3828, 2.4062, 2.8594, 1.1016, 2.6875, 1.1797, 3.0469, 2.2344, 2.4219, 0.7812, 0.5273, 1.3594, 0.7734, 1.2812, 2.9531, 2.0469, 0.9141, 1.1484, 2.0625, 2.4375, 1.4922, 2.3281, 2.2812, 1.0391, 1.375, 2.3906, 2.2344, 1.3516, 2.2656, 3.8438, 1.3672, 2.875, 2.1719, 1.2266, 1.5312, 1.5547, 2.125, 1.2656, 4.5625, 1.0781, 2.3906, 1.5547, 1.9062, 1.5781, 1.9219, 2.8906, 1.5938, 1.6719, 0.7969, 2.375, 0.7383, 2.2344, 1.4922, 2.3594, 2.3438, 1.7188, 2.5312, 2.1094, 4.4062, 1.3359, 1.7812, 0.8086, 1.875, 1.7656, 2.0781, 1.1797, 2.4219, 1.1328, 1.5391, 2.5938, 1.5547, 2.25, 1.7969, 0.9883, 2.1562, 1.1328, 1.1562, 1.375, 2.2812, 4.6562, 2.2344, 4.1875, 2.125, 2.5625, 1.4766, 1.0078, 1.4844, 1.4922, 1.5234, 1.5391, 1.8516, 0.9961, 1.2188, 2.0938, 1.2031, 0.9805, 1.5391, 2.4062, 1.6016, 2.3125, 2.25, 0.8789, 1.7266, 1.5547, 1.2578, 1.4062, 1.3281, 1.3203, 3.4844, 1.3594, 1.2656, 1.5547, 1.1016, 4.0625, 1.7031, 1.6953, 0.7773, 0.5703, 3.3906, 2.25, 2.1562, 1.3359, 2.0625, 1.4375, 2.125, 2.0469, 1.8125, 1.0234, 1.5312, 2.7188, 1.8359, 2.0781, 1.875, 1.8516, 2.2188, 1.7578, 0.9883, 2.2969, 0.8047, 1.6641, 1.7344, 1.8516, 1.6484, 2.375, 1.2734, 2.25, 1.4297, 1.8203, 2.0625, 1.7344, 2.5469, 1.1719, 2.375, 1.7188, 0.7695, 2.2812, 2.0156, 0.5078, 4.5625, 1.7188, 1.9844, 0.7188, 1.3359, 0.9531, 1.3516, 0.8633, 2.1406, 1.7109, 0.8086, 2.3594, 1.6484, 2.4688, 2.1719, 2.6875, 2.2969, 1.4609, 1.3594, 0.75, 1.5938, 0.9453, 1.8906, 1.0391, 0.7578, 0.8203, 1.0781, 0.7305, 2.2656, 1.5391, 0.9688, 0.707, 0.7852, 1.8906, 1.4922, 2.25, 2.1562, 0.8047, 2.875, 0.8906, 2.3281, 1.8516, 1.9844, 1.5234, 0.8164, 1.2734, 2.5312, 4.875, 1.4688, 2.3594, 2.2031, 2.3594, 1.9219, 2.4375, 1.2422, 1.8047, 2.1562, 1.7812, 1.2656, 1.6094, 1.7578, 1.4219, 0.7266, 1.8906, 1.0, 1.7812, 2.0, 1.9609, 1.1875, 3.75, 1.7344, 1.5703, 2.5469, 1.8047, 0.9922, 2.0469, 2.2344, 1.3047, 2.0625, 2.2812, 2.3281, 3.4844, 1.5, 1.5078, 2.2188, 1.5938, 1.6484, 2.4688, 1.3047, 2.3438, 2.1875, 1.7578, 1.3047, 2.5312, 1.2422, 1.2422, 2.3125, 4.7188, 1.3438, 1.0625, 1.2188, 1.8828, 0.8594, 1.3281, 0.9922, 2.7812, 2.2188, 0.8242, 2.7031, 2.1562, 1.7344, 1.3672, 1.3438, 1.4453, 1.6875, 0.8633, 1.25, 1.7109, 2.4219, 1.625, 0.9648, 1.9531, 1.9844, 1.4688, 1.3438, 2.2812, 1.3672, 2.5625, 4.375, 1.6641, 1.7656, 0.8164, 2.375, 2.3906, 2.0312, 1.3203, 1.2109, 1.2812, 2.0, 2.1719, 2.8906, 1.4297, 1.7344, 2.1406, 4.0312, 2.0, 1.8203, 1.1953, 2.0, 1.625, 1.8594, 1.4219, 1.1016, 1.5547, 3.3438, 1.3828, 2.25, 1.7969, 0.9766, 1.0625, 2.375, 2.0625, 1.0703, 1.6719, 1.8594, 0.3594, 1.4062, 1.4531, 1.8828, 1.5625, 1.2734, 1.7109, 2.2812, 1.2422, 1.6484, 2.0156, 1.3047, 1.2969, 0.6406, 0.7344, 0.8281, 2.2969, 0.7578, 1.8672, 1.3281, 0.7617, 2.25, 1.125, 1.2109, 2.0469, 1.5312, 1.1875, 0.7539, 2.3906, 0.8633, 2.4531, 1.2578, 1.5859, 2.3125, 2.4844, 0.7812, 1.4766, 0.7734, 1.6484, 0.9062, 1.6328, 1.5469, 0.5898, 1.7422, 1.0312, 1.2812, 1.5703, 2.3906, 1.3672, 1.7266]
|
Chapter 3/chapter_3_loss_vizualisation.ipynb
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
Chapter 3/chapter_3_padding_vizualisations.ipynb
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
Chapter 3/constraint_padding.png
ADDED
|
Git LFS Details
|
Chapter 3/create_fig_3_1.py
ADDED
|
@@ -0,0 +1,220 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""
|
| 2 |
+
iMessage-style visualization for VLM dataset samples.
|
| 3 |
+
User = right-aligned blue bubbles with white text.
|
| 4 |
+
Assistant = left-aligned light-gray bubbles with dark text.
|
| 5 |
+
Image inlined after the first user message (right-aligned).
|
| 6 |
+
Max 2 turns. Generates 10 images.
|
| 7 |
+
"""
|
| 8 |
+
|
| 9 |
+
import sys, os
|
| 10 |
+
from datasets import load_dataset
|
| 11 |
+
from PIL import Image, ImageDraw, ImageFont
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
# ── Fonts ────────────────────────────────────────────────────────────
|
| 15 |
+
|
| 16 |
+
def get_font(size):
|
| 17 |
+
for name in [
|
| 18 |
+
"/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf",
|
| 19 |
+
"DejaVuSans.ttf",
|
| 20 |
+
"Arial.ttf",
|
| 21 |
+
"/System/Library/Fonts/Helvetica.ttc",
|
| 22 |
+
]:
|
| 23 |
+
try:
|
| 24 |
+
return ImageFont.truetype(name, size)
|
| 25 |
+
except (OSError, IOError):
|
| 26 |
+
continue
|
| 27 |
+
return ImageFont.load_default()
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
FONT = get_font(20)
|
| 31 |
+
WIDTH = 600
|
| 32 |
+
PAD = 16
|
| 33 |
+
BUBBLE_PAD_H = 16
|
| 34 |
+
BUBBLE_PAD_V = 12
|
| 35 |
+
LINE_HEIGHT = 28
|
| 36 |
+
BUBBLE_MAX_W = int(WIDTH * 0.72)
|
| 37 |
+
BUBBLE_RADIUS = 18
|
| 38 |
+
GAP = 10
|
| 39 |
+
BG_COLOR = "#F2F2F7"
|
| 40 |
+
|
| 41 |
+
USER_BG = "#007AFF"
|
| 42 |
+
USER_TEXT = "#FFFFFF"
|
| 43 |
+
ASST_BG = "#E9E9EB"
|
| 44 |
+
ASST_TEXT = "#000000"
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
# ── Data helpers ─────────────────────────────────────────────────────
|
| 48 |
+
|
| 49 |
+
def extract_sample_image(sample):
|
| 50 |
+
if "image" in sample and sample["image"] is not None:
|
| 51 |
+
return sample["image"]
|
| 52 |
+
if "images" in sample and sample["images"]:
|
| 53 |
+
return sample["images"][0]
|
| 54 |
+
raise KeyError("No image found in sample.")
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
def normalize_conversations(sample, max_turns=2):
|
| 58 |
+
raw = sample.get("texts", [])
|
| 59 |
+
normalized = []
|
| 60 |
+
for msg in raw:
|
| 61 |
+
if not isinstance(msg, dict):
|
| 62 |
+
continue
|
| 63 |
+
if "role" in msg and "content" in msg:
|
| 64 |
+
normalized.append({"role": msg["role"], "content": msg["content"]})
|
| 65 |
+
continue
|
| 66 |
+
if "user" in msg:
|
| 67 |
+
normalized.append({"role": "user", "content": str(msg["user"])})
|
| 68 |
+
if "assistant" in msg:
|
| 69 |
+
normalized.append({"role": "assistant", "content": str(msg["assistant"])})
|
| 70 |
+
|
| 71 |
+
turns, current = [], []
|
| 72 |
+
for msg in normalized:
|
| 73 |
+
current.append(msg)
|
| 74 |
+
if msg["role"] == "assistant":
|
| 75 |
+
turns.append(current)
|
| 76 |
+
current = []
|
| 77 |
+
if current:
|
| 78 |
+
turns.append(current)
|
| 79 |
+
turns = turns[:max_turns]
|
| 80 |
+
return [msg for turn in turns for msg in turn]
|
| 81 |
+
|
| 82 |
+
|
| 83 |
+
# ── Drawing helpers ──────────────────────────────────────────────────
|
| 84 |
+
|
| 85 |
+
def wrap_text(text, font, max_width, draw):
|
| 86 |
+
lines = []
|
| 87 |
+
for paragraph in text.split("\n"):
|
| 88 |
+
if not paragraph.strip():
|
| 89 |
+
lines.append("")
|
| 90 |
+
continue
|
| 91 |
+
words = paragraph.split()
|
| 92 |
+
cur = ""
|
| 93 |
+
for w in words:
|
| 94 |
+
test = f"{cur} {w}".strip()
|
| 95 |
+
bbox = draw.textbbox((0, 0), test, font=font)
|
| 96 |
+
if bbox[2] - bbox[0] > max_width:
|
| 97 |
+
if cur:
|
| 98 |
+
lines.append(cur)
|
| 99 |
+
cur = w
|
| 100 |
+
else:
|
| 101 |
+
cur = test
|
| 102 |
+
if cur:
|
| 103 |
+
lines.append(cur)
|
| 104 |
+
return lines
|
| 105 |
+
|
| 106 |
+
|
| 107 |
+
def measure_bubble(lines):
|
| 108 |
+
tmp = Image.new("RGB", (1, 1))
|
| 109 |
+
d = ImageDraw.Draw(tmp)
|
| 110 |
+
text_w = 0
|
| 111 |
+
for line in lines:
|
| 112 |
+
bbox = d.textbbox((0, 0), line, font=FONT)
|
| 113 |
+
text_w = max(text_w, bbox[2] - bbox[0])
|
| 114 |
+
text_h = len(lines) * LINE_HEIGHT
|
| 115 |
+
bw = text_w + 2 * BUBBLE_PAD_H
|
| 116 |
+
bh = text_h + 2 * BUBBLE_PAD_V
|
| 117 |
+
return bw, bh
|
| 118 |
+
|
| 119 |
+
|
| 120 |
+
def round_image_corners(img, radius):
|
| 121 |
+
w, h = img.size
|
| 122 |
+
mask = Image.new("L", (w, h), 0)
|
| 123 |
+
d = ImageDraw.Draw(mask)
|
| 124 |
+
d.rounded_rectangle([0, 0, w, h], radius=radius, fill=255)
|
| 125 |
+
result = img.copy().convert("RGBA")
|
| 126 |
+
result.putalpha(mask)
|
| 127 |
+
return result
|
| 128 |
+
|
| 129 |
+
|
| 130 |
+
# ── Main renderer ────────────────────────────────────────────────────
|
| 131 |
+
|
| 132 |
+
def render_imessage(sample_image, conversations, filename="imessage_style.png"):
|
| 133 |
+
tmp = Image.new("RGB", (WIDTH, 100))
|
| 134 |
+
tmp_draw = ImageDraw.Draw(tmp)
|
| 135 |
+
inner_text_w = BUBBLE_MAX_W - 2 * BUBBLE_PAD_H
|
| 136 |
+
|
| 137 |
+
items = []
|
| 138 |
+
image_inserted = False
|
| 139 |
+
|
| 140 |
+
for msg in conversations:
|
| 141 |
+
role = msg["role"]
|
| 142 |
+
text = msg["content"]
|
| 143 |
+
|
| 144 |
+
text = text.strip()
|
| 145 |
+
lines = wrap_text(text, FONT, inner_text_w, tmp_draw)
|
| 146 |
+
bw, bh = measure_bubble(lines)
|
| 147 |
+
items.append(("bubble", (role, lines, bw, bh), bh + GAP))
|
| 148 |
+
|
| 149 |
+
# Insert image right after the first user message
|
| 150 |
+
if role == "user" and not image_inserted:
|
| 151 |
+
img_w, img_h = sample_image.size
|
| 152 |
+
max_img_w = BUBBLE_MAX_W
|
| 153 |
+
max_img_h = 320
|
| 154 |
+
scale = min(max_img_w / img_w, max_img_h / img_h, 1.0)
|
| 155 |
+
new_w = int(img_w * scale)
|
| 156 |
+
new_h = int(img_h * scale)
|
| 157 |
+
scaled = sample_image.resize((new_w, new_h), Image.LANCZOS)
|
| 158 |
+
items.append(("image", scaled, new_h + 8))
|
| 159 |
+
image_inserted = True
|
| 160 |
+
|
| 161 |
+
total_h = PAD
|
| 162 |
+
for _, _, h in items:
|
| 163 |
+
total_h += h + GAP
|
| 164 |
+
total_h += PAD
|
| 165 |
+
|
| 166 |
+
canvas = Image.new("RGB", (WIDTH, total_h), BG_COLOR)
|
| 167 |
+
draw = ImageDraw.Draw(canvas)
|
| 168 |
+
y = PAD
|
| 169 |
+
|
| 170 |
+
for item_type, data, h in items:
|
| 171 |
+
if item_type == "image":
|
| 172 |
+
scaled = data
|
| 173 |
+
sw, sh = scaled.size
|
| 174 |
+
ix = WIDTH - PAD - sw
|
| 175 |
+
rounded = round_image_corners(scaled, BUBBLE_RADIUS)
|
| 176 |
+
canvas.paste(rounded, (ix, y), rounded)
|
| 177 |
+
y += sh + GAP + 4
|
| 178 |
+
|
| 179 |
+
elif item_type == "bubble":
|
| 180 |
+
role, lines, bw, bh = data
|
| 181 |
+
is_user = role == "user"
|
| 182 |
+
bg = USER_BG if is_user else ASST_BG
|
| 183 |
+
text_color = USER_TEXT if is_user else ASST_TEXT
|
| 184 |
+
bx = (WIDTH - PAD - bw) if is_user else PAD
|
| 185 |
+
|
| 186 |
+
draw.rounded_rectangle(
|
| 187 |
+
[bx, y, bx + bw, y + bh],
|
| 188 |
+
radius=BUBBLE_RADIUS,
|
| 189 |
+
fill=bg,
|
| 190 |
+
)
|
| 191 |
+
|
| 192 |
+
ty = y + BUBBLE_PAD_V
|
| 193 |
+
for line in lines:
|
| 194 |
+
draw.text((bx + BUBBLE_PAD_H, ty), line, fill=text_color, font=FONT)
|
| 195 |
+
ty += LINE_HEIGHT
|
| 196 |
+
y += bh + GAP
|
| 197 |
+
|
| 198 |
+
canvas.save(filename)
|
| 199 |
+
print(f"Saved {filename} ({WIDTH}x{total_h})")
|
| 200 |
+
return canvas
|
| 201 |
+
|
| 202 |
+
|
| 203 |
+
# ── Generate 10 images ───────────────────────────────────────────────
|
| 204 |
+
|
| 205 |
+
dataset = load_dataset("HuggingFaceM4/FineVisionMax", split="train", streaming=True)
|
| 206 |
+
|
| 207 |
+
count = 0
|
| 208 |
+
for sample in dataset:
|
| 209 |
+
if count >= 50:
|
| 210 |
+
break
|
| 211 |
+
try:
|
| 212 |
+
sample_image = extract_sample_image(sample).convert("RGB")
|
| 213 |
+
conversations = normalize_conversations(sample, max_turns=2)
|
| 214 |
+
except (KeyError, ValueError):
|
| 215 |
+
continue
|
| 216 |
+
render_imessage(sample_image, conversations, filename=f"imessage_style_{count:02d}.png")
|
| 217 |
+
count += 1
|
| 218 |
+
|
| 219 |
+
print(f"Done — saved {count} images.")
|
| 220 |
+
os._exit(0)
|
Chapter 3/create_fig_3_2.tsx
ADDED
|
@@ -0,0 +1,221 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import { useState, useRef, useCallback } from "react";
|
| 2 |
+
|
| 3 |
+
const TOKENS = ["the", "a", "cat", "sun", "is", "big", "ran", "to", "red", "it"];
|
| 4 |
+
const CORRECT = 3;
|
| 5 |
+
const lowLoss = [0.04, 0.02, 0.06, 0.52, 0.04, 0.06, 0.03, 0.05, 0.12, 0.06];
|
| 6 |
+
const highLoss = [0.05, 0.03, 0.22, 0.04, 0.06, 0.28, 0.08, 0.05, 0.12, 0.07];
|
| 7 |
+
const ce = (dist, c) => -Math.log(Math.max(dist[c], 1e-9));
|
| 8 |
+
|
| 9 |
+
const SCALE = 2;
|
| 10 |
+
|
| 11 |
+
function drawChart(ctx, ox, oy, title, dist, subtitle, W, H) {
|
| 12 |
+
const PAD = { top: 100, right: 20, bottom: 64, left: 46 };
|
| 13 |
+
const pw = W - PAD.left - PAD.right;
|
| 14 |
+
const ph = H - PAD.top - PAD.bottom;
|
| 15 |
+
const n = dist.length;
|
| 16 |
+
const gap = 5;
|
| 17 |
+
const bw = (pw - gap * (n - 1)) / n;
|
| 18 |
+
const maxP = 0.55;
|
| 19 |
+
const loss = ce(dist, CORRECT);
|
| 20 |
+
const isLow = loss < 1;
|
| 21 |
+
|
| 22 |
+
ctx.save();
|
| 23 |
+
ctx.translate(ox, oy);
|
| 24 |
+
|
| 25 |
+
// Title
|
| 26 |
+
ctx.font = `bold ${15 * SCALE}px Inter, -apple-system, system-ui, sans-serif`;
|
| 27 |
+
ctx.fillStyle = "#222";
|
| 28 |
+
ctx.textAlign = "center";
|
| 29 |
+
ctx.fillText(title, W / 2, 24 * SCALE);
|
| 30 |
+
|
| 31 |
+
// Y grid + labels
|
| 32 |
+
ctx.font = `${10 * SCALE}px Inter, -apple-system, system-ui, sans-serif`;
|
| 33 |
+
ctx.textAlign = "right";
|
| 34 |
+
for (const t of [0, 0.1, 0.2, 0.3, 0.4, 0.5]) {
|
| 35 |
+
const y = PAD.top + ph * (1 - t / maxP);
|
| 36 |
+
ctx.strokeStyle = "#e0e0e0";
|
| 37 |
+
ctx.lineWidth = 1;
|
| 38 |
+
ctx.beginPath(); ctx.moveTo(PAD.left, y); ctx.lineTo(W - PAD.right, y); ctx.stroke();
|
| 39 |
+
ctx.fillStyle = "#999";
|
| 40 |
+
ctx.fillText(t === 0 ? "0" : `${(t * 100).toFixed(0)}%`, PAD.left - 8, y + 4 * SCALE);
|
| 41 |
+
}
|
| 42 |
+
|
| 43 |
+
// Baseline
|
| 44 |
+
ctx.strokeStyle = "#ccc";
|
| 45 |
+
ctx.lineWidth = 1.5;
|
| 46 |
+
ctx.beginPath(); ctx.moveTo(PAD.left, PAD.top + ph); ctx.lineTo(W - PAD.right, PAD.top + ph); ctx.stroke();
|
| 47 |
+
|
| 48 |
+
// Bars
|
| 49 |
+
for (let i = 0; i < n; i++) {
|
| 50 |
+
const p = dist[i];
|
| 51 |
+
const x = PAD.left + i * (bw + gap);
|
| 52 |
+
const barH = (p / maxP) * ph;
|
| 53 |
+
const y = PAD.top + ph - barH;
|
| 54 |
+
const isC = i === CORRECT;
|
| 55 |
+
|
| 56 |
+
const r = 4 * SCALE;
|
| 57 |
+
const bx = x, by = y, bww = bw, bhh = barH;
|
| 58 |
+
|
| 59 |
+
ctx.fillStyle = isC ? (isLow ? "#22a861" : "#d63031") : "#b0b5bc";
|
| 60 |
+
ctx.globalAlpha = isC ? 1 : 0.5;
|
| 61 |
+
ctx.beginPath();
|
| 62 |
+
ctx.moveTo(bx + r, by);
|
| 63 |
+
ctx.lineTo(bx + bww - r, by);
|
| 64 |
+
ctx.quadraticCurveTo(bx + bww, by, bx + bww, by + r);
|
| 65 |
+
ctx.lineTo(bx + bww, by + bhh);
|
| 66 |
+
ctx.lineTo(bx, by + bhh);
|
| 67 |
+
ctx.lineTo(bx, by + r);
|
| 68 |
+
ctx.quadraticCurveTo(bx, by, bx + r, by);
|
| 69 |
+
ctx.closePath();
|
| 70 |
+
ctx.fill();
|
| 71 |
+
ctx.globalAlpha = 1;
|
| 72 |
+
|
| 73 |
+
// Percentage on correct bar
|
| 74 |
+
if (isC) {
|
| 75 |
+
ctx.font = `bold ${10 * SCALE}px Inter, sans-serif`;
|
| 76 |
+
ctx.fillStyle = "#333";
|
| 77 |
+
ctx.textAlign = "center";
|
| 78 |
+
ctx.fillText(`${(p * 100).toFixed(0)}%`, x + bw / 2, y - 14 * SCALE);
|
| 79 |
+
}
|
| 80 |
+
|
| 81 |
+
// Token label
|
| 82 |
+
ctx.font = `${isC ? "bold " : ""}${10.5 * SCALE}px 'SF Mono', 'Fira Code', monospace`;
|
| 83 |
+
ctx.fillStyle = isC ? "#222" : "#999";
|
| 84 |
+
ctx.textAlign = "center";
|
| 85 |
+
ctx.fillText(TOKENS[i], x + bw / 2, PAD.top + ph + 18 * SCALE);
|
| 86 |
+
|
| 87 |
+
// "correct" marker
|
| 88 |
+
if (isC) {
|
| 89 |
+
ctx.font = `${8 * SCALE}px Inter, sans-serif`;
|
| 90 |
+
ctx.fillStyle = "#777";
|
| 91 |
+
ctx.letterSpacing = "1px";
|
| 92 |
+
ctx.fillText("correct", x + bw / 2, PAD.top + ph + 32 * SCALE);
|
| 93 |
+
|
| 94 |
+
// Arrow
|
| 95 |
+
const ax = x + bw / 2;
|
| 96 |
+
const ay1 = PAD.top + ph + 36 * SCALE;
|
| 97 |
+
const ay2 = ay1 + 7 * SCALE;
|
| 98 |
+
ctx.strokeStyle = "#aaa";
|
| 99 |
+
ctx.lineWidth = 1.5;
|
| 100 |
+
ctx.beginPath(); ctx.moveTo(ax, ay1); ctx.lineTo(ax, ay2); ctx.stroke();
|
| 101 |
+
ctx.fillStyle = "#aaa";
|
| 102 |
+
ctx.beginPath();
|
| 103 |
+
ctx.moveTo(ax - 4 * SCALE, ay2);
|
| 104 |
+
ctx.lineTo(ax + 4 * SCALE, ay2);
|
| 105 |
+
ctx.lineTo(ax, ay2 + 5 * SCALE);
|
| 106 |
+
ctx.closePath();
|
| 107 |
+
ctx.fill();
|
| 108 |
+
}
|
| 109 |
+
}
|
| 110 |
+
|
| 111 |
+
// Pill
|
| 112 |
+
const pillW = 200 * SCALE;
|
| 113 |
+
const pillH = 30 * SCALE;
|
| 114 |
+
const pillX = (W - pillW) / 2;
|
| 115 |
+
const pillY = H - 10 * SCALE;
|
| 116 |
+
|
| 117 |
+
ctx.fillStyle = isLow ? "#dff5e8" : "#fce4e4";
|
| 118 |
+
ctx.beginPath();
|
| 119 |
+
const pr = pillH / 2;
|
| 120 |
+
ctx.moveTo(pillX + pr, pillY);
|
| 121 |
+
ctx.lineTo(pillX + pillW - pr, pillY);
|
| 122 |
+
ctx.quadraticCurveTo(pillX + pillW, pillY, pillX + pillW, pillY + pr);
|
| 123 |
+
ctx.quadraticCurveTo(pillX + pillW, pillY + pillH, pillX + pillW - pr, pillY + pillH);
|
| 124 |
+
ctx.lineTo(pillX + pr, pillY + pillH);
|
| 125 |
+
ctx.quadraticCurveTo(pillX, pillY + pillH, pillX, pillY + pr);
|
| 126 |
+
ctx.quadraticCurveTo(pillX, pillY, pillX + pr, pillY);
|
| 127 |
+
ctx.closePath();
|
| 128 |
+
ctx.fill();
|
| 129 |
+
|
| 130 |
+
ctx.font = `bold ${11 * SCALE}px Inter, sans-serif`;
|
| 131 |
+
ctx.fillStyle = isLow ? "#1e8449" : "#b71c1c";
|
| 132 |
+
ctx.textAlign = "center";
|
| 133 |
+
ctx.fillText(`${subtitle} · loss = ${loss.toFixed(2)}`, W / 2, pillY + pillH / 2 + 4 * SCALE);
|
| 134 |
+
|
| 135 |
+
ctx.restore();
|
| 136 |
+
}
|
| 137 |
+
|
| 138 |
+
export default function App() {
|
| 139 |
+
const canvasRef = useRef(null);
|
| 140 |
+
const [rendered, setRendered] = useState(false);
|
| 141 |
+
|
| 142 |
+
const chartW = 320 * SCALE;
|
| 143 |
+
const chartH = 280 * SCALE;
|
| 144 |
+
const totalGap = 50 * SCALE;
|
| 145 |
+
const padX = 30 * SCALE;
|
| 146 |
+
const padTop = 50 * SCALE;
|
| 147 |
+
const padBot = 30 * SCALE;
|
| 148 |
+
const totalW = padX * 2 + chartW * 2 + totalGap;
|
| 149 |
+
const totalH = padTop + chartH + padBot;
|
| 150 |
+
|
| 151 |
+
const draw = useCallback((canvas) => {
|
| 152 |
+
if (!canvas || rendered) return;
|
| 153 |
+
canvasRef.current = canvas;
|
| 154 |
+
const ctx = canvas.getContext("2d");
|
| 155 |
+
|
| 156 |
+
// White background
|
| 157 |
+
ctx.fillStyle = "#ffffff";
|
| 158 |
+
ctx.fillRect(0, 0, totalW, totalH);
|
| 159 |
+
|
| 160 |
+
// Header
|
| 161 |
+
ctx.font = `${13 * SCALE}px Inter, -apple-system, system-ui, sans-serif`;
|
| 162 |
+
ctx.fillStyle = "#222";
|
| 163 |
+
ctx.textAlign = "center";
|
| 164 |
+
ctx.fillText("Predicted probability distribution over vocabulary", totalW / 2, 22 * SCALE);
|
| 165 |
+
|
| 166 |
+
ctx.font = `${11 * SCALE}px 'SF Mono', 'Fira Code', monospace`;
|
| 167 |
+
ctx.fillStyle = "#222";
|
| 168 |
+
ctx.fillText('The model predicts the next token after: "the __ rises"', totalW / 2, 38 * SCALE);
|
| 169 |
+
|
| 170 |
+
drawChart(ctx, padX, padTop, "Low surprise", lowLoss, "Confident & correct", chartW, chartH);
|
| 171 |
+
drawChart(ctx, padX + chartW + totalGap, padTop, "High surprise", highLoss, "Spread & wrong", chartW, chartH);
|
| 172 |
+
|
| 173 |
+
setRendered(true);
|
| 174 |
+
}, [rendered, totalW, totalH]);
|
| 175 |
+
|
| 176 |
+
const download = () => {
|
| 177 |
+
const c = canvasRef.current;
|
| 178 |
+
if (!c) return;
|
| 179 |
+
const link = document.createElement("a");
|
| 180 |
+
link.download = "cross-entropy-loss.png";
|
| 181 |
+
link.href = c.toDataURL("image/png");
|
| 182 |
+
link.click();
|
| 183 |
+
};
|
| 184 |
+
|
| 185 |
+
return (
|
| 186 |
+
<div style={{
|
| 187 |
+
display: "flex", flexDirection: "column", alignItems: "center", justifyContent: "center",
|
| 188 |
+
minHeight: "100vh", background: "#f5f5f5",
|
| 189 |
+
fontFamily: "'Inter', -apple-system, system-ui, sans-serif",
|
| 190 |
+
padding: 24, gap: 20,
|
| 191 |
+
}}>
|
| 192 |
+
<canvas
|
| 193 |
+
ref={draw}
|
| 194 |
+
width={totalW}
|
| 195 |
+
height={totalH}
|
| 196 |
+
style={{
|
| 197 |
+
width: totalW / SCALE,
|
| 198 |
+
height: totalH / SCALE,
|
| 199 |
+
borderRadius: 8,
|
| 200 |
+
boxShadow: "0 2px 16px rgba(0,0,0,0.07)",
|
| 201 |
+
}}
|
| 202 |
+
/>
|
| 203 |
+
<button
|
| 204 |
+
onClick={download}
|
| 205 |
+
style={{
|
| 206 |
+
padding: "10px 28px",
|
| 207 |
+
borderRadius: 8,
|
| 208 |
+
border: "none",
|
| 209 |
+
background: "#222",
|
| 210 |
+
color: "#fff",
|
| 211 |
+
fontSize: 14,
|
| 212 |
+
fontWeight: 600,
|
| 213 |
+
cursor: "pointer",
|
| 214 |
+
letterSpacing: 0.3,
|
| 215 |
+
}}
|
| 216 |
+
>
|
| 217 |
+
Download PNG
|
| 218 |
+
</button>
|
| 219 |
+
</div>
|
| 220 |
+
);
|
| 221 |
+
}
|
Chapter 3/greedy_knapsack_packing.png
ADDED
|
Git LFS Details
|
Chapter 3/knapsack_packing.png
ADDED
|
Git LFS Details
|
Chapter 3/loss_comparison.png
ADDED
|
Git LFS Details
|
Chapter 3/loss_plot_first_5000_steps.png
ADDED
|
Git LFS Details
|
Chapter 3/minimal_vlm_training.ipynb
ADDED
|
@@ -0,0 +1,2307 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cells": [
|
| 3 |
+
{
|
| 4 |
+
"cell_type": "code",
|
| 5 |
+
"execution_count": 1,
|
| 6 |
+
"metadata": {
|
| 7 |
+
"colab": {
|
| 8 |
+
"base_uri": "https://localhost:8080/",
|
| 9 |
+
"height": 223,
|
| 10 |
+
"referenced_widgets": [
|
| 11 |
+
"6af6b0d5fb5942f0874062514fecdac9",
|
| 12 |
+
"aaceb8bb8d144e62a7fc19dde30e981b",
|
| 13 |
+
"9a9026e81b974c7ba30f92d2e7d5f16f",
|
| 14 |
+
"81bebd074665482b9af8e6d39b6b171a",
|
| 15 |
+
"c1891461614e45189d47fa3948da6808",
|
| 16 |
+
"db773254517a428684eb679771212f2c",
|
| 17 |
+
"5cb1c4db303b4b6aaaabc7cf88e975d5",
|
| 18 |
+
"edcb5218b9bc4e9d9da375eb93de2829",
|
| 19 |
+
"59c2f3d1128a45ebbc2f1405ae42751a",
|
| 20 |
+
"8de705db750048ff890e74afcf97f964",
|
| 21 |
+
"032dcdea26234490a141913d7e361a15",
|
| 22 |
+
"f05122bb2187477ab3f5181938c86a21",
|
| 23 |
+
"e74fce8c74c548d285e41413679a080d",
|
| 24 |
+
"3686ac8683c745ee8bc5638d1e962c54",
|
| 25 |
+
"9049575e797c4bffa86129fd9394f8fd",
|
| 26 |
+
"64363326d62641f3aee10f317d92140e",
|
| 27 |
+
"6baf2c3128d1425c8f979ed703b0fc6b",
|
| 28 |
+
"2f69578d14bb4ad6967e00931f611f63",
|
| 29 |
+
"b391ef77f8ed4d4f9ccd41ace6a80075",
|
| 30 |
+
"7f01a3c81bb34621aa7f2744722de6bf",
|
| 31 |
+
"6d97eea8e68948698db477acaa7ff4f6",
|
| 32 |
+
"dbbd386fb13b4dbcae40301c1bbbbf90"
|
| 33 |
+
]
|
| 34 |
+
},
|
| 35 |
+
"id": "DzFaTkPrpVse",
|
| 36 |
+
"outputId": "90b58e6c-046d-4982-c94e-8372cc44d6f5"
|
| 37 |
+
},
|
| 38 |
+
"outputs": [
|
| 39 |
+
{
|
| 40 |
+
"name": "stderr",
|
| 41 |
+
"output_type": "stream",
|
| 42 |
+
"text": [
|
| 43 |
+
"Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.\n"
|
| 44 |
+
]
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"data": {
|
| 48 |
+
"application/vnd.jupyter.widget-view+json": {
|
| 49 |
+
"model_id": "da8eb18c731c47e59a4aeff1800b16a5",
|
| 50 |
+
"version_major": 2,
|
| 51 |
+
"version_minor": 0
|
| 52 |
+
},
|
| 53 |
+
"text/plain": [
|
| 54 |
+
"README.md: 0.00B [00:00, ?B/s]"
|
| 55 |
+
]
|
| 56 |
+
},
|
| 57 |
+
"metadata": {},
|
| 58 |
+
"output_type": "display_data"
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"data": {
|
| 62 |
+
"application/vnd.jupyter.widget-view+json": {
|
| 63 |
+
"model_id": "678bdcb6f77a42e8b5e50f86267094fd",
|
| 64 |
+
"version_major": 2,
|
| 65 |
+
"version_minor": 0
|
| 66 |
+
},
|
| 67 |
+
"text/plain": [
|
| 68 |
+
"Resolving data files: 0%| | 0/10000 [00:00<?, ?it/s]"
|
| 69 |
+
]
|
| 70 |
+
},
|
| 71 |
+
"metadata": {},
|
| 72 |
+
"output_type": "display_data"
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"name": "stdout",
|
| 76 |
+
"output_type": "stream",
|
| 77 |
+
"text": [
|
| 78 |
+
"{'images': [<PIL.PngImagePlugin.PngImageFile image mode=RGB size=1506x1102 at 0xEC954E9D5600>], 'texts': [{'user': 'What is the primary key of the Equipment entity?', 'assistant': 'The primary key of the Equipment entity is EquipmentID, which uniquely identifies each piece of equipment in the diagram.\\nAnswer: EquipmentID'}, {'user': 'What is the relationship between Equipment and Maintenance Record?', 'assistant': 'The relationship between Equipment and Maintenance Record is labeled as \"Maintained By\", indicating that one piece of equipment can have multiple maintenance records associated with it.\\nAnswer: Maintained By'}, {'user': 'How many relationships does the Supplier entity have?', 'assistant': 'The Supplier entity has two relationships: \"Supplied By\" with Equipment and \"Delivered By\" with Logistics Company, indicating it serves as a midpoint in both flows.\\nAnswer: 2'}, {'user': 'What is the cardinality of the relationship between Equipment and Supplier?', 'assistant': 'The cardinality of the relationship between Equipment and Supplier is M:1, which means many pieces of equipment can be supplied by one supplier.\\nAnswer: M:1'}, {'user': 'Which entity is connected to Logistics Company through the Delivered By relationship?', 'assistant': 'The entity connected to Logistics Company through the Delivered By relationship is Supplier, indicating that the supplier relies on the logistics company for delivering equipment.\\nAnswer: Supplier'}, {'user': 'How many attributes does the Maintenance Record entity have?', 'assistant': 'The Maintenance Record entity has five attributes: RecordID, Date, Description, ServiceCost, and ServiceProvider, detailing the specifics of each maintenance entry.\\nAnswer: 5'}, {'user': 'What type of diagram is represented in this illustration?', 'assistant': 'This illustration represents an Entity Relationship Diagram, which shows entities, their attributes, and how they are related to one another within a system, particularly for equipment maintenance and service records.\\nAnswer: Entity Relationship Diagram'}, {'user': 'Which attribute of Supplier is related to contact information?', 'assistant': 'The contact information attribute related to Supplier is ContactNumber which provides the phone number for communication.\\nAnswer: ContactNumber'}, {'user': 'Who maintains the Equipment records?', 'assistant': 'The Equipment records are maintained through the relationship labeled \"Maintained By\" with the Maintenance Record, indicating a direct link between equipment details and their records of maintenance.\\nAnswer: Maintenance Record'}], 'source': 'CoSyn-400k-diagram', 'relevance_ratings': [5, 5, 5, 5, 5, 5, 5, 5, 2], 'relevance_min': 2, 'visual_dependency_ratings': [1, 2, 4, 4, 4, 4, 5, 2, 2], 'visual_dependency_min': 1, 'image_correspondence_ratings': [3, 5, 4, 4, 5, 5, 5, 3, 5], 'image_correspondence_min': 3, 'formatting_ratings': [4, 4, 4, 4, 4, 5, 4, 4, 4], 'formatting_min': 4}\n"
|
| 79 |
+
]
|
| 80 |
+
}
|
| 81 |
+
],
|
| 82 |
+
"source": [
|
| 83 |
+
"from datasets import load_dataset\n",
|
| 84 |
+
"\n",
|
| 85 |
+
"dataset = load_dataset(\"HuggingFaceM4/FineVisionMax\", split=\"train\", streaming=True)\n",
|
| 86 |
+
"print(next(iter(dataset)))"
|
| 87 |
+
]
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"cell_type": "code",
|
| 91 |
+
"execution_count": 30,
|
| 92 |
+
"metadata": {
|
| 93 |
+
"id": "AU4DEPrKpeAX"
|
| 94 |
+
},
|
| 95 |
+
"outputs": [],
|
| 96 |
+
"source": [
|
| 97 |
+
"import torch\n",
|
| 98 |
+
"import torch.nn as nn\n",
|
| 99 |
+
"from transformers import AutoModel, AutoProcessor, AutoTokenizer, AutoModelForCausalLM\n",
|
| 100 |
+
"\n",
|
| 101 |
+
"class VisionLanguageModel(nn.Module):\n",
|
| 102 |
+
" def __init__(self, vision_encoder_ckpt, language_model_ckpt, modality_input_dim=768, modality_output_dim=576):\n",
|
| 103 |
+
" super().__init__()\n",
|
| 104 |
+
" self.vision_encoder = AutoModel.from_pretrained(vision_encoder_ckpt).vision_model # only take vision backbone\n",
|
| 105 |
+
" self.vision_processor = AutoProcessor.from_pretrained(vision_encoder_ckpt)\n",
|
| 106 |
+
" self.modality_projector = nn.Linear(modality_input_dim, modality_output_dim, bias=False)\n",
|
| 107 |
+
" self.tokenizer = AutoTokenizer.from_pretrained(language_model_ckpt)\n",
|
| 108 |
+
" self.llm = AutoModelForCausalLM.from_pretrained(language_model_ckpt)\n",
|
| 109 |
+
"\n",
|
| 110 |
+
" def forward(self, text, image, labels=None):\n",
|
| 111 |
+
" processed_img = self.vision_processor(images=[image], return_tensors=\"pt\").to(self.llm.device)\n",
|
| 112 |
+
" image_embd = self.vision_encoder(**processed_img).last_hidden_state\n",
|
| 113 |
+
" image_embd = self.modality_projector(image_embd).to(dtype=self.llm.dtype) # match LLM dtype\n",
|
| 114 |
+
"\n",
|
| 115 |
+
" input_ids = self.tokenizer(text, return_tensors=\"pt\").input_ids.to(self.llm.device)\n",
|
| 116 |
+
" token_embd = self.llm.model.embed_tokens(input_ids)\n",
|
| 117 |
+
"\n",
|
| 118 |
+
" combined_embd = torch.cat((image_embd, token_embd), dim=1) # Concatenate image embeddings to token embeddings\n",
|
| 119 |
+
"\n",
|
| 120 |
+
" logits = self.llm(inputs_embeds=combined_embd).logits\n",
|
| 121 |
+
"\n",
|
| 122 |
+
" shift_logits = logits[:, image_embd.size(1):-1, :].contiguous()\n",
|
| 123 |
+
" shift_labels = input_ids[:, 1:].contiguous()\n",
|
| 124 |
+
"\n",
|
| 125 |
+
" loss = nn.functional.cross_entropy(shift_logits.reshape(-1, shift_logits.size(-1)), shift_labels.reshape(-1))\n",
|
| 126 |
+
"\n",
|
| 127 |
+
" return logits, loss"
|
| 128 |
+
]
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"cell_type": "code",
|
| 132 |
+
"execution_count": 31,
|
| 133 |
+
"metadata": {
|
| 134 |
+
"colab": {
|
| 135 |
+
"base_uri": "https://localhost:8080/",
|
| 136 |
+
"height": 81,
|
| 137 |
+
"referenced_widgets": [
|
| 138 |
+
"059fc34bbe9c40a0a81c5640003ccc47",
|
| 139 |
+
"3cc6112c0037496a9337382c1d16a3d6",
|
| 140 |
+
"f4fcf974faad42a08fb6863763eedb81",
|
| 141 |
+
"d6cc3547f86c4629a8e8e53592167b43",
|
| 142 |
+
"c04849b1932040188a760a2ebabc91ec",
|
| 143 |
+
"5ae3364706d04df0afcd0e941c09a01a",
|
| 144 |
+
"f09cbd8806bc41808b5594cfb16c1a83",
|
| 145 |
+
"18111ed761e146698786fb4a5d0c7051",
|
| 146 |
+
"1e523935a83d434c976f10797cbe2286",
|
| 147 |
+
"9db3a6aae167458b8e3c6b785c25f225",
|
| 148 |
+
"cd7b84b3c61e49d093df411ccfbc3bb0",
|
| 149 |
+
"8d851590df7c4cf3a9fc293d03befa40",
|
| 150 |
+
"aa9c7dbcf7b04f339fce5bee0d24c98d",
|
| 151 |
+
"fba6cbf72ad54e748031d11669b3e31a",
|
| 152 |
+
"296ba3ca9f34496ba8edfeee4c76a5d2",
|
| 153 |
+
"cc1e631bd9154962bcb42eb6fc234303",
|
| 154 |
+
"a65cbec1df084615915186797051a9e3",
|
| 155 |
+
"4d95c6fe1c434425b1368f07952e250a",
|
| 156 |
+
"891d641ad7634c9aa5186589f2aba958",
|
| 157 |
+
"aa902a918c79418ea4b4516fb5673fa4",
|
| 158 |
+
"d8ee4430ef8c47bc96d6d7ae2060f9ad",
|
| 159 |
+
"667ff0bc56f94ab582252968d3dca89b"
|
| 160 |
+
]
|
| 161 |
+
},
|
| 162 |
+
"id": "V3lGxumXq9VD",
|
| 163 |
+
"outputId": "ff0b55d3-b0a3-45ce-f377-8c460684af8e"
|
| 164 |
+
},
|
| 165 |
+
"outputs": [
|
| 166 |
+
{
|
| 167 |
+
"data": {
|
| 168 |
+
"application/vnd.jupyter.widget-view+json": {
|
| 169 |
+
"model_id": "0b811e26e7524bb4baba26f713332410",
|
| 170 |
+
"version_major": 2,
|
| 171 |
+
"version_minor": 0
|
| 172 |
+
},
|
| 173 |
+
"text/plain": [
|
| 174 |
+
"Loading weights: 0%| | 0/408 [00:00<?, ?it/s]"
|
| 175 |
+
]
|
| 176 |
+
},
|
| 177 |
+
"metadata": {},
|
| 178 |
+
"output_type": "display_data"
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"data": {
|
| 182 |
+
"application/vnd.jupyter.widget-view+json": {
|
| 183 |
+
"model_id": "f5ca88c6eb384283a7a5d13ed1474514",
|
| 184 |
+
"version_major": 2,
|
| 185 |
+
"version_minor": 0
|
| 186 |
+
},
|
| 187 |
+
"text/plain": [
|
| 188 |
+
"Loading weights: 0%| | 0/272 [00:00<?, ?it/s]"
|
| 189 |
+
]
|
| 190 |
+
},
|
| 191 |
+
"metadata": {},
|
| 192 |
+
"output_type": "display_data"
|
| 193 |
+
}
|
| 194 |
+
],
|
| 195 |
+
"source": [
|
| 196 |
+
"vision_encoder_ckpt = \"google/siglip2-base-patch16-256\"\n",
|
| 197 |
+
"language_model_ckpt = \"HuggingFaceTB/SmolLM2-135M-Instruct\"\n",
|
| 198 |
+
"\n",
|
| 199 |
+
"vlm = VisionLanguageModel(vision_encoder_ckpt, language_model_ckpt).to(\"cuda\")"
|
| 200 |
+
]
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"cell_type": "code",
|
| 204 |
+
"execution_count": 28,
|
| 205 |
+
"metadata": {
|
| 206 |
+
"colab": {
|
| 207 |
+
"base_uri": "https://localhost:8080/"
|
| 208 |
+
},
|
| 209 |
+
"id": "lux-u_uCrVBJ",
|
| 210 |
+
"outputId": "79411cab-c7f2-4d67-fceb-e9b14c9611e2"
|
| 211 |
+
},
|
| 212 |
+
"outputs": [
|
| 213 |
+
{
|
| 214 |
+
"name": "stdout",
|
| 215 |
+
"output_type": "stream",
|
| 216 |
+
"text": [
|
| 217 |
+
"torch.Size([1, 256, 576])\n"
|
| 218 |
+
]
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"data": {
|
| 222 |
+
"text/plain": [
|
| 223 |
+
"(tensor([[[15.0000, 6.1875, 9.3750, ..., 6.9375, 10.7500, 2.0469],\n",
|
| 224 |
+
" [22.8750, 11.6875, 14.8750, ..., 17.8750, 19.1250, 16.8750],\n",
|
| 225 |
+
" [20.7500, 5.8125, 7.5312, ..., 12.5000, 15.8750, 10.5000],\n",
|
| 226 |
+
" ...,\n",
|
| 227 |
+
" [13.9375, 2.1250, 3.8906, ..., 10.4375, 12.1250, 8.1250],\n",
|
| 228 |
+
" [13.6250, 3.0625, 6.1875, ..., 8.0625, 12.3750, 5.6562],\n",
|
| 229 |
+
" [14.3125, 5.4062, 10.1250, ..., 9.1250, 12.5625, 6.8750]]],\n",
|
| 230 |
+
" device='cuda:0', dtype=torch.bfloat16, grad_fn=<UnsafeViewBackward0>),\n",
|
| 231 |
+
" tensor(4.8125, device='cuda:0', dtype=torch.bfloat16,\n",
|
| 232 |
+
" grad_fn=<NllLossBackward0>))"
|
| 233 |
+
]
|
| 234 |
+
},
|
| 235 |
+
"execution_count": 28,
|
| 236 |
+
"metadata": {},
|
| 237 |
+
"output_type": "execute_result"
|
| 238 |
+
}
|
| 239 |
+
],
|
| 240 |
+
"source": [
|
| 241 |
+
"from transformers.image_utils import load_image\n",
|
| 242 |
+
"\n",
|
| 243 |
+
"image = load_image(\"https://huggingface.co/datasets/merve/coco/resolve/main/val2017/000000000285.jpg\")\n",
|
| 244 |
+
"text = \"What does this say?\"\n",
|
| 245 |
+
"vlm(text, image)"
|
| 246 |
+
]
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"cell_type": "code",
|
| 250 |
+
"execution_count": 10,
|
| 251 |
+
"metadata": {
|
| 252 |
+
"id": "jAMaabaNr7JS"
|
| 253 |
+
},
|
| 254 |
+
"outputs": [],
|
| 255 |
+
"source": [
|
| 256 |
+
"sample = next(iter(dataset))\n",
|
| 257 |
+
"\n",
|
| 258 |
+
"def apply_chat_template_to_sample(sample):\n",
|
| 259 |
+
" messages_list = []\n",
|
| 260 |
+
" for data_dict in sample['texts']:\n",
|
| 261 |
+
" messages_list.append({'role': 'user', 'content': data_dict['user']})\n",
|
| 262 |
+
" messages_list.append({'role': 'assistant', 'content': data_dict['assistant']})\n",
|
| 263 |
+
" text = vlm.tokenizer.apply_chat_template(messages_list, tokenize=False)\n",
|
| 264 |
+
" return text\n",
|
| 265 |
+
"\n",
|
| 266 |
+
"text = apply_chat_template_to_sample(sample)\n"
|
| 267 |
+
]
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"cell_type": "code",
|
| 271 |
+
"execution_count": 11,
|
| 272 |
+
"metadata": {
|
| 273 |
+
"colab": {
|
| 274 |
+
"base_uri": "https://localhost:8080/",
|
| 275 |
+
"height": 141
|
| 276 |
+
},
|
| 277 |
+
"id": "280WsrwW1Ets",
|
| 278 |
+
"outputId": "2024d829-7240-4861-ccd9-4ce04dbd390d"
|
| 279 |
+
},
|
| 280 |
+
"outputs": [
|
| 281 |
+
{
|
| 282 |
+
"data": {
|
| 283 |
+
"text/plain": [
|
| 284 |
+
"{'images': [<PIL.PngImagePlugin.PngImageFile image mode=RGB size=1506x1102>],\n",
|
| 285 |
+
" 'texts': [{'user': 'What is the primary key of the Equipment entity?',\n",
|
| 286 |
+
" 'assistant': 'The primary key of the Equipment entity is EquipmentID, which uniquely identifies each piece of equipment in the diagram.\\nAnswer: EquipmentID'},\n",
|
| 287 |
+
" {'user': 'What is the relationship between Equipment and Maintenance Record?',\n",
|
| 288 |
+
" 'assistant': 'The relationship between Equipment and Maintenance Record is labeled as \"Maintained By\", indicating that one piece of equipment can have multiple maintenance records associated with it.\\nAnswer: Maintained By'},\n",
|
| 289 |
+
" {'user': 'How many relationships does the Supplier entity have?',\n",
|
| 290 |
+
" 'assistant': 'The Supplier entity has two relationships: \"Supplied By\" with Equipment and \"Delivered By\" with Logistics Company, indicating it serves as a midpoint in both flows.\\nAnswer: 2'},\n",
|
| 291 |
+
" {'user': 'What is the cardinality of the relationship between Equipment and Supplier?',\n",
|
| 292 |
+
" 'assistant': 'The cardinality of the relationship between Equipment and Supplier is M:1, which means many pieces of equipment can be supplied by one supplier.\\nAnswer: M:1'},\n",
|
| 293 |
+
" {'user': 'Which entity is connected to Logistics Company through the Delivered By relationship?',\n",
|
| 294 |
+
" 'assistant': 'The entity connected to Logistics Company through the Delivered By relationship is Supplier, indicating that the supplier relies on the logistics company for delivering equipment.\\nAnswer: Supplier'},\n",
|
| 295 |
+
" {'user': 'How many attributes does the Maintenance Record entity have?',\n",
|
| 296 |
+
" 'assistant': 'The Maintenance Record entity has five attributes: RecordID, Date, Description, ServiceCost, and ServiceProvider, detailing the specifics of each maintenance entry.\\nAnswer: 5'},\n",
|
| 297 |
+
" {'user': 'What type of diagram is represented in this illustration?',\n",
|
| 298 |
+
" 'assistant': 'This illustration represents an Entity Relationship Diagram, which shows entities, their attributes, and how they are related to one another within a system, particularly for equipment maintenance and service records.\\nAnswer: Entity Relationship Diagram'},\n",
|
| 299 |
+
" {'user': 'Which attribute of Supplier is related to contact information?',\n",
|
| 300 |
+
" 'assistant': 'The contact information attribute related to Supplier is ContactNumber which provides the phone number for communication.\\nAnswer: ContactNumber'},\n",
|
| 301 |
+
" {'user': 'Who maintains the Equipment records?',\n",
|
| 302 |
+
" 'assistant': 'The Equipment records are maintained through the relationship labeled \"Maintained By\" with the Maintenance Record, indicating a direct link between equipment details and their records of maintenance.\\nAnswer: Maintenance Record'}],\n",
|
| 303 |
+
" 'source': 'CoSyn-400k-diagram',\n",
|
| 304 |
+
" 'relevance_ratings': [5, 5, 5, 5, 5, 5, 5, 5, 2],\n",
|
| 305 |
+
" 'relevance_min': 2,\n",
|
| 306 |
+
" 'visual_dependency_ratings': [1, 2, 4, 4, 4, 4, 5, 2, 2],\n",
|
| 307 |
+
" 'visual_dependency_min': 1,\n",
|
| 308 |
+
" 'image_correspondence_ratings': [3, 5, 4, 4, 5, 5, 5, 3, 5],\n",
|
| 309 |
+
" 'image_correspondence_min': 3,\n",
|
| 310 |
+
" 'formatting_ratings': [4, 4, 4, 4, 4, 5, 4, 4, 4],\n",
|
| 311 |
+
" 'formatting_min': 4}"
|
| 312 |
+
]
|
| 313 |
+
},
|
| 314 |
+
"execution_count": 11,
|
| 315 |
+
"metadata": {},
|
| 316 |
+
"output_type": "execute_result"
|
| 317 |
+
}
|
| 318 |
+
],
|
| 319 |
+
"source": [
|
| 320 |
+
"sample"
|
| 321 |
+
]
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"cell_type": "code",
|
| 325 |
+
"execution_count": 12,
|
| 326 |
+
"metadata": {},
|
| 327 |
+
"outputs": [
|
| 328 |
+
{
|
| 329 |
+
"data": {
|
| 330 |
+
"text/plain": [
|
| 331 |
+
"'<|im_start|>system\\nYou are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>\\n<|im_start|>user\\nWhat is the primary key of the Equipment entity?<|im_end|>\\n<|im_start|>assistant\\nThe primary key of the Equipment entity is EquipmentID, which uniquely identifies each piece of equipment in the diagram.\\nAnswer: EquipmentID<|im_end|>\\n<|im_start|>user\\nWhat is the relationship between Equipment and Maintenance Record?<|im_end|>\\n<|im_start|>assistant\\nThe relationship between Equipment and Maintenance Record is labeled as \"Maintained By\", indicating that one piece of equipment can have multiple maintenance records associated with it.\\nAnswer: Maintained By<|im_end|>\\n<|im_start|>user\\nHow many relationships does the Supplier entity have?<|im_end|>\\n<|im_start|>assistant\\nThe Supplier entity has two relationships: \"Supplied By\" with Equipment and \"Delivered By\" with Logistics Company, indicating it serves as a midpoint in both flows.\\nAnswer: 2<|im_end|>\\n<|im_start|>user\\nWhat is the cardinality of the relationship between Equipment and Supplier?<|im_end|>\\n<|im_start|>assistant\\nThe cardinality of the relationship between Equipment and Supplier is M:1, which means many pieces of equipment can be supplied by one supplier.\\nAnswer: M:1<|im_end|>\\n<|im_start|>user\\nWhich entity is connected to Logistics Company through the Delivered By relationship?<|im_end|>\\n<|im_start|>assistant\\nThe entity connected to Logistics Company through the Delivered By relationship is Supplier, indicating that the supplier relies on the logistics company for delivering equipment.\\nAnswer: Supplier<|im_end|>\\n<|im_start|>user\\nHow many attributes does the Maintenance Record entity have?<|im_end|>\\n<|im_start|>assistant\\nThe Maintenance Record entity has five attributes: RecordID, Date, Description, ServiceCost, and ServiceProvider, detailing the specifics of each maintenance entry.\\nAnswer: 5<|im_end|>\\n<|im_start|>user\\nWhat type of diagram is represented in this illustration?<|im_end|>\\n<|im_start|>assistant\\nThis illustration represents an Entity Relationship Diagram, which shows entities, their attributes, and how they are related to one another within a system, particularly for equipment maintenance and service records.\\nAnswer: Entity Relationship Diagram<|im_end|>\\n<|im_start|>user\\nWhich attribute of Supplier is related to contact information?<|im_end|>\\n<|im_start|>assistant\\nThe contact information attribute related to Supplier is ContactNumber which provides the phone number for communication.\\nAnswer: ContactNumber<|im_end|>\\n<|im_start|>user\\nWho maintains the Equipment records?<|im_end|>\\n<|im_start|>assistant\\nThe Equipment records are maintained through the relationship labeled \"Maintained By\" with the Maintenance Record, indicating a direct link between equipment details and their records of maintenance.\\nAnswer: Maintenance Record<|im_end|>\\n'"
|
| 332 |
+
]
|
| 333 |
+
},
|
| 334 |
+
"execution_count": 12,
|
| 335 |
+
"metadata": {},
|
| 336 |
+
"output_type": "execute_result"
|
| 337 |
+
}
|
| 338 |
+
],
|
| 339 |
+
"source": [
|
| 340 |
+
"text"
|
| 341 |
+
]
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"cell_type": "code",
|
| 345 |
+
"execution_count": 13,
|
| 346 |
+
"metadata": {
|
| 347 |
+
"id": "PFNRdFus5ZfM"
|
| 348 |
+
},
|
| 349 |
+
"outputs": [
|
| 350 |
+
{
|
| 351 |
+
"name": "stdout",
|
| 352 |
+
"output_type": "stream",
|
| 353 |
+
"text": [
|
| 354 |
+
"torch.Size([1, 256, 576])\n"
|
| 355 |
+
]
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"data": {
|
| 359 |
+
"text/plain": [
|
| 360 |
+
"(tensor([[[18.7500, 11.0000, 15.1250, ..., 12.6875, 15.3750, 8.4375],\n",
|
| 361 |
+
" [19.5000, 10.0625, 14.5625, ..., 17.5000, 16.5000, 12.5625],\n",
|
| 362 |
+
" [16.7500, 9.6875, 13.6250, ..., 11.4375, 14.0000, 10.3750],\n",
|
| 363 |
+
" ...,\n",
|
| 364 |
+
" [17.8750, 25.5000, 32.7500, ..., 16.0000, 11.5000, 9.9375],\n",
|
| 365 |
+
" [24.6250, 27.6250, 29.8750, ..., 13.6875, 13.0000, 11.3750],\n",
|
| 366 |
+
" [13.5000, 31.8750, 31.6250, ..., 10.7500, 8.9375, 8.5625]]],\n",
|
| 367 |
+
" device='cuda:0', dtype=torch.bfloat16, grad_fn=<UnsafeViewBackward0>),\n",
|
| 368 |
+
" tensor(1.9062, device='cuda:0', dtype=torch.bfloat16,\n",
|
| 369 |
+
" grad_fn=<NllLossBackward0>))"
|
| 370 |
+
]
|
| 371 |
+
},
|
| 372 |
+
"execution_count": 13,
|
| 373 |
+
"metadata": {},
|
| 374 |
+
"output_type": "execute_result"
|
| 375 |
+
}
|
| 376 |
+
],
|
| 377 |
+
"source": [
|
| 378 |
+
"vlm(text, sample['images'])"
|
| 379 |
+
]
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"cell_type": "code",
|
| 383 |
+
"execution_count": 32,
|
| 384 |
+
"metadata": {
|
| 385 |
+
"colab": {
|
| 386 |
+
"base_uri": "https://localhost:8080/"
|
| 387 |
+
},
|
| 388 |
+
"id": "AT9eCHhU2cSe",
|
| 389 |
+
"outputId": "e1e8e953-c40b-4177-ff62-1c49ad2cafdd"
|
| 390 |
+
},
|
| 391 |
+
"outputs": [
|
| 392 |
+
{
|
| 393 |
+
"name": "stdout",
|
| 394 |
+
"output_type": "stream",
|
| 395 |
+
"text": [
|
| 396 |
+
"step 100 | loss 1.9766\n",
|
| 397 |
+
"step 200 | loss 2.7969\n",
|
| 398 |
+
"step 300 | loss 2.0938\n",
|
| 399 |
+
"step 400 | loss 5.1875\n",
|
| 400 |
+
"step 500 | loss 2.5625\n",
|
| 401 |
+
"step 600 | loss 2.5000\n",
|
| 402 |
+
"step 700 | loss 1.3516\n",
|
| 403 |
+
"step 800 | loss 1.0000\n",
|
| 404 |
+
"step 900 | loss 1.5625\n",
|
| 405 |
+
"step 1000 | loss 1.7812\n",
|
| 406 |
+
"step 1100 | loss 2.4688\n",
|
| 407 |
+
"step 1200 | loss 1.6250\n",
|
| 408 |
+
"step 1300 | loss 2.1094\n",
|
| 409 |
+
"step 1400 | loss 1.1641\n",
|
| 410 |
+
"step 1500 | loss 0.9219\n",
|
| 411 |
+
"step 1600 | loss 1.5000\n",
|
| 412 |
+
"step 1700 | loss 2.2969\n",
|
| 413 |
+
"step 1800 | loss 1.4219\n",
|
| 414 |
+
"step 1900 | loss 2.5781\n",
|
| 415 |
+
"step 2000 | loss 1.8828\n",
|
| 416 |
+
"step 2100 | loss 0.6445\n",
|
| 417 |
+
"step 2200 | loss 0.7930\n",
|
| 418 |
+
"step 2300 | loss 1.5078\n",
|
| 419 |
+
"step 2400 | loss 2.2969\n",
|
| 420 |
+
"step 2500 | loss 2.2812\n",
|
| 421 |
+
"step 2600 | loss 1.7031\n",
|
| 422 |
+
"step 2700 | loss 1.9531\n",
|
| 423 |
+
"step 2800 | loss 2.4375\n",
|
| 424 |
+
"step 2900 | loss 0.7578\n",
|
| 425 |
+
"step 3000 | loss 3.6562\n",
|
| 426 |
+
"step 3100 | loss 4.1250\n",
|
| 427 |
+
"step 3200 | loss 1.8281\n",
|
| 428 |
+
"step 3300 | loss 2.3125\n",
|
| 429 |
+
"step 3400 | loss 1.6406\n",
|
| 430 |
+
"step 3500 | loss 2.4375\n",
|
| 431 |
+
"step 3600 | loss 1.8047\n",
|
| 432 |
+
"step 3700 | loss 1.9531\n",
|
| 433 |
+
"step 3800 | loss 2.0000\n",
|
| 434 |
+
"step 3900 | loss 1.7812\n",
|
| 435 |
+
"step 4000 | loss 1.2891\n",
|
| 436 |
+
"step 4100 | loss 1.8516\n",
|
| 437 |
+
"step 4200 | loss 1.8438\n",
|
| 438 |
+
"step 4300 | loss 0.8086\n",
|
| 439 |
+
"step 4400 | loss 2.0156\n",
|
| 440 |
+
"step 4500 | loss 2.1719\n",
|
| 441 |
+
"step 4600 | loss 2.1875\n",
|
| 442 |
+
"step 4700 | loss 2.2500\n",
|
| 443 |
+
"step 4800 | loss 2.4688\n",
|
| 444 |
+
"step 4900 | loss 1.4453\n",
|
| 445 |
+
"step 5000 | loss 1.3672\n"
|
| 446 |
+
]
|
| 447 |
+
}
|
| 448 |
+
],
|
| 449 |
+
"source": [
|
| 450 |
+
"import json\n",
|
| 451 |
+
"import torch.optim as optim\n",
|
| 452 |
+
"optimizer = optim.AdamW(vlm.parameters(), lr=1e-5)\n",
|
| 453 |
+
"\n",
|
| 454 |
+
"step = 0\n",
|
| 455 |
+
"losses = []\n",
|
| 456 |
+
"for sample in dataset:\n",
|
| 457 |
+
" text = apply_chat_template_to_sample(sample)\n",
|
| 458 |
+
" if len(text) > 3000 or len(sample['images']) != 1:\n",
|
| 459 |
+
" continue\n",
|
| 460 |
+
" images = [sample['images'][0].convert('RGB')]\n",
|
| 461 |
+
" step += 1\n",
|
| 462 |
+
"\n",
|
| 463 |
+
" optimizer.zero_grad()\n",
|
| 464 |
+
" logits, loss = vlm(text, images)\n",
|
| 465 |
+
" losses.append(loss)\n",
|
| 466 |
+
"\n",
|
| 467 |
+
" loss.backward()\n",
|
| 468 |
+
" optimizer.step()\n",
|
| 469 |
+
"\n",
|
| 470 |
+
" if step%100 == 0:\n",
|
| 471 |
+
" print(f\"step {step} | loss {loss.item():.4f}\")\n",
|
| 472 |
+
"\n",
|
| 473 |
+
" if step > 5000:\n",
|
| 474 |
+
" json.dump([round(l.item(), 4) for l in losses], open(\"chapter_3_loss.json\", \"w\"))\n",
|
| 475 |
+
" break"
|
| 476 |
+
]
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"cell_type": "code",
|
| 480 |
+
"execution_count": 1,
|
| 481 |
+
"metadata": {
|
| 482 |
+
"colab": {
|
| 483 |
+
"base_uri": "https://localhost:8080/",
|
| 484 |
+
"height": 226
|
| 485 |
+
},
|
| 486 |
+
"id": "1nLOpTCs4q4F",
|
| 487 |
+
"outputId": "ea6a05a6-cfc3-483a-ac22-1ca31f7be664"
|
| 488 |
+
},
|
| 489 |
+
"outputs": [
|
| 490 |
+
{
|
| 491 |
+
"name": "stderr",
|
| 492 |
+
"output_type": "stream",
|
| 493 |
+
"text": [
|
| 494 |
+
"Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.\n"
|
| 495 |
+
]
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"data": {
|
| 499 |
+
"application/vnd.jupyter.widget-view+json": {
|
| 500 |
+
"model_id": "58cd5e73761a416a851346fc23dff3cb",
|
| 501 |
+
"version_major": 2,
|
| 502 |
+
"version_minor": 0
|
| 503 |
+
},
|
| 504 |
+
"text/plain": [
|
| 505 |
+
"Resolving data files: 0%| | 0/10000 [00:00<?, ?it/s]"
|
| 506 |
+
]
|
| 507 |
+
},
|
| 508 |
+
"metadata": {},
|
| 509 |
+
"output_type": "display_data"
|
| 510 |
+
}
|
| 511 |
+
],
|
| 512 |
+
"source": [
|
| 513 |
+
"from datasets import load_dataset\n",
|
| 514 |
+
"from transformers import AutoTokenizer\n",
|
| 515 |
+
"\n",
|
| 516 |
+
"dataset = load_dataset(\"HuggingFaceM4/FineVisionMax\", split=\"train\", streaming=True)\n",
|
| 517 |
+
"tokenizer = AutoTokenizer.from_pretrained(\"HuggingFaceTB/SmolLM2-135M-Instruct\")\n",
|
| 518 |
+
"\n",
|
| 519 |
+
"def tokenize_sample(sample):\n",
|
| 520 |
+
" messages_list = []\n",
|
| 521 |
+
" for data_dict in sample['texts']:\n",
|
| 522 |
+
" messages_list.append({'role': 'user', 'content': data_dict['user']})\n",
|
| 523 |
+
" messages_list.append({'role': 'assistant', 'content': data_dict['assistant']})\n",
|
| 524 |
+
" tokens = tokenizer.apply_chat_template(messages_list, tokenize=True)\n",
|
| 525 |
+
" return tokens['input_ids']\n",
|
| 526 |
+
"\n",
|
| 527 |
+
"max_length = 2048\n",
|
| 528 |
+
"curr_sample = []\n",
|
| 529 |
+
"batch = []\n",
|
| 530 |
+
"max_batch_size = 8\n",
|
| 531 |
+
"for sample in dataset:\n",
|
| 532 |
+
" tokens = tokenize_sample(sample)\n",
|
| 533 |
+
" if len(tokens) > max_length:\n",
|
| 534 |
+
" continue # Skip samples that are too long\n",
|
| 535 |
+
" if len(curr_sample) + len(tokens) < max_length:\n",
|
| 536 |
+
" curr_sample.extend(tokens)\n",
|
| 537 |
+
" else:\n",
|
| 538 |
+
" if len(batch) == max_batch_size:\n",
|
| 539 |
+
" break # We've reached the max batch size, so we need to break\n",
|
| 540 |
+
" batch.append(curr_sample)\n",
|
| 541 |
+
" curr_sample = tokens"
|
| 542 |
+
]
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"cell_type": "code",
|
| 546 |
+
"execution_count": null,
|
| 547 |
+
"metadata": {
|
| 548 |
+
"id": "TQ7S--kRFAc-"
|
| 549 |
+
},
|
| 550 |
+
"outputs": [],
|
| 551 |
+
"source": []
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"cell_type": "code",
|
| 555 |
+
"execution_count": 2,
|
| 556 |
+
"metadata": {
|
| 557 |
+
"id": "a81sI4oFFA5d"
|
| 558 |
+
},
|
| 559 |
+
"outputs": [],
|
| 560 |
+
"source": [
|
| 561 |
+
"import torch\n",
|
| 562 |
+
"import torch.nn as nn\n",
|
| 563 |
+
"from transformers import AutoModel, AutoProcessor, AutoTokenizer, AutoModelForCausalLM\n",
|
| 564 |
+
"\n",
|
| 565 |
+
"class VisionLanguageModel(nn.Module):\n",
|
| 566 |
+
" def __init__(self, vision_encoder_ckpt, language_model_ckpt, tokenizer, modality_input_dim=768, modality_output_dim=576):\n",
|
| 567 |
+
" super().__init__()\n",
|
| 568 |
+
" self.vision_encoder = AutoModel.from_pretrained(vision_encoder_ckpt).vision_model\n",
|
| 569 |
+
" self.vision_processor = AutoProcessor.from_pretrained(vision_encoder_ckpt)\n",
|
| 570 |
+
" self.modality_projector = nn.Linear(modality_input_dim, modality_output_dim, bias=False)\n",
|
| 571 |
+
" self.tokenizer = tokenizer # AutoTokenizer.from_pretrained(language_model_ckpt, extra_special_tokens={\"image_token\": \"<|image|>\"})\n",
|
| 572 |
+
" self.llm = AutoModelForCausalLM.from_pretrained(language_model_ckpt)\n",
|
| 573 |
+
" self.llm.resize_token_embeddings(len(self.tokenizer)) # Resize the LLM's token embeddings to match the tokenizer's new vocab size\n",
|
| 574 |
+
"\n",
|
| 575 |
+
" def _replace_img_tokens_with_embd(self, input_ids, token_embd, image_embd):\n",
|
| 576 |
+
" \"\"\"\n",
|
| 577 |
+
" Replace every image-token placeholder in `input_ids` with the corresponding slice\n",
|
| 578 |
+
" from `image_embd`. Supports an arbitrary number of image-token placeholders per sample.\n",
|
| 579 |
+
" The first example in the batch might have 2 images and the second none.\n",
|
| 580 |
+
" \"\"\"\n",
|
| 581 |
+
" # Clone the original embeddings to avoid in-place issues\n",
|
| 582 |
+
" updated_token_embd = token_embd.clone()\n",
|
| 583 |
+
"\n",
|
| 584 |
+
" # Build a mask of all image-token positions: shape [B, T_seq]\n",
|
| 585 |
+
" mask = (input_ids == self.tokenizer.image_token_id)\n",
|
| 586 |
+
" updated_token_embd[mask] = image_embd.view(-1, image_embd.size(-1)).to(updated_token_embd.dtype) # torch flattens before assigning\n",
|
| 587 |
+
"\n",
|
| 588 |
+
" return updated_token_embd\n",
|
| 589 |
+
"\n",
|
| 590 |
+
" def forward(self, input_ids, image):\n",
|
| 591 |
+
" processed_img = self.vision_processor(images=[image], return_tensors=\"pt\").to(self.llm.device)\n",
|
| 592 |
+
" image_embd = self.vision_encoder(**processed_img).last_hidden_state\n",
|
| 593 |
+
" image_embd = self.modality_projector(image_embd)\n",
|
| 594 |
+
"\n",
|
| 595 |
+
" token_embd = self.llm.model.embed_tokens(input_ids)\n",
|
| 596 |
+
" combined_embd = self._replace_img_tokens_with_embd(input_ids, token_embd, image_embd)\n",
|
| 597 |
+
"\n",
|
| 598 |
+
" logits = self.llm(inputs_embeds=combined_embd).logits\n",
|
| 599 |
+
"\n",
|
| 600 |
+
" # 1. Create labels from the input_ids\n",
|
| 601 |
+
" labels = input_ids.clone()\n",
|
| 602 |
+
"\n",
|
| 603 |
+
" # 2. We don't want to compute loss on image tokens.\n",
|
| 604 |
+
" # Set them to the ignore_index (which is pad_token_id).\n",
|
| 605 |
+
" labels[labels == self.tokenizer.image_token_id] = self.tokenizer.pad_token_id\n",
|
| 606 |
+
"\n",
|
| 607 |
+
" # 3. Standard next-token-prediction shifting:\n",
|
| 608 |
+
" # The logit at index `i` predicts the token at index `i+1`\n",
|
| 609 |
+
" shift_logits = logits[..., :-1, :].contiguous()\n",
|
| 610 |
+
" shift_labels = labels[..., 1:].contiguous()\n",
|
| 611 |
+
"\n",
|
| 612 |
+
" # 4. Compute loss, ignoring all tokens marked with pad_token_id\n",
|
| 613 |
+
" # (this now includes both padding AND image tokens)\n",
|
| 614 |
+
" loss = nn.functional.cross_entropy(\n",
|
| 615 |
+
" shift_logits.reshape(-1, shift_logits.size(-1)),\n",
|
| 616 |
+
" shift_labels.reshape(-1),\n",
|
| 617 |
+
" ignore_index=self.tokenizer.pad_token_id\n",
|
| 618 |
+
" )\n",
|
| 619 |
+
"\n",
|
| 620 |
+
" return logits, loss"
|
| 621 |
+
]
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"cell_type": "code",
|
| 625 |
+
"execution_count": 19,
|
| 626 |
+
"metadata": {
|
| 627 |
+
"id": "rdfs-3Q-Gf9u"
|
| 628 |
+
},
|
| 629 |
+
"outputs": [
|
| 630 |
+
{
|
| 631 |
+
"data": {
|
| 632 |
+
"application/vnd.jupyter.widget-view+json": {
|
| 633 |
+
"model_id": "d3e4c194362d4f9dbd2861ab057850c9",
|
| 634 |
+
"version_major": 2,
|
| 635 |
+
"version_minor": 0
|
| 636 |
+
},
|
| 637 |
+
"text/plain": [
|
| 638 |
+
"Resolving data files: 0%| | 0/10000 [00:00<?, ?it/s]"
|
| 639 |
+
]
|
| 640 |
+
},
|
| 641 |
+
"metadata": {},
|
| 642 |
+
"output_type": "display_data"
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"data": {
|
| 646 |
+
"application/vnd.jupyter.widget-view+json": {
|
| 647 |
+
"model_id": "003b90cf7584444398ae733d2b8cd547",
|
| 648 |
+
"version_major": 2,
|
| 649 |
+
"version_minor": 0
|
| 650 |
+
},
|
| 651 |
+
"text/plain": [
|
| 652 |
+
"Loading weights: 0%| | 0/408 [00:00<?, ?it/s]"
|
| 653 |
+
]
|
| 654 |
+
},
|
| 655 |
+
"metadata": {},
|
| 656 |
+
"output_type": "display_data"
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"data": {
|
| 660 |
+
"application/vnd.jupyter.widget-view+json": {
|
| 661 |
+
"model_id": "d882c51b85d74a3c9a3426bddee36b8e",
|
| 662 |
+
"version_major": 2,
|
| 663 |
+
"version_minor": 0
|
| 664 |
+
},
|
| 665 |
+
"text/plain": [
|
| 666 |
+
"Loading weights: 0%| | 0/272 [00:00<?, ?it/s]"
|
| 667 |
+
]
|
| 668 |
+
},
|
| 669 |
+
"metadata": {},
|
| 670 |
+
"output_type": "display_data"
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"name": "stderr",
|
| 674 |
+
"output_type": "stream",
|
| 675 |
+
"text": [
|
| 676 |
+
"Token indices sequence length is longer than the specified maximum sequence length for this model (28674 > 8192). Running this sequence through the model will result in indexing errors\n"
|
| 677 |
+
]
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"name": "stdout",
|
| 681 |
+
"output_type": "stream",
|
| 682 |
+
"text": [
|
| 683 |
+
"step 100 | loss 1.4531 | smoothed loss: 1.5258\n",
|
| 684 |
+
"The last 100 steps took: 91.58 seconds.\n",
|
| 685 |
+
"step 200 | loss 1.4609 | smoothed loss: 1.4172\n",
|
| 686 |
+
"The last 100 steps took: 93.71 seconds.\n",
|
| 687 |
+
"step 300 | loss 1.5703 | smoothed loss: 1.3359\n",
|
| 688 |
+
"The last 100 steps took: 93.15 seconds.\n",
|
| 689 |
+
"step 400 | loss 1.5156 | smoothed loss: 1.3262\n",
|
| 690 |
+
"The last 100 steps took: 93.01 seconds.\n",
|
| 691 |
+
"step 500 | loss 1.0859 | smoothed loss: 1.2203\n",
|
| 692 |
+
"The last 100 steps took: 73.74 seconds.\n",
|
| 693 |
+
"step 600 | loss 1.3281 | smoothed loss: 1.3719\n",
|
| 694 |
+
"The last 100 steps took: 96.48 seconds.\n",
|
| 695 |
+
"step 700 | loss 0.6992 | smoothed loss: 1.2375\n",
|
| 696 |
+
"The last 100 steps took: 92.60 seconds.\n",
|
| 697 |
+
"step 800 | loss 1.2656 | smoothed loss: 1.3758\n",
|
| 698 |
+
"The last 100 steps took: 95.97 seconds.\n",
|
| 699 |
+
"step 900 | loss 1.1484 | smoothed loss: 1.2699\n",
|
| 700 |
+
"The last 100 steps took: 84.72 seconds.\n",
|
| 701 |
+
"step 1000 | loss 1.5000 | smoothed loss: 1.3703\n",
|
| 702 |
+
"The last 100 steps took: 75.39 seconds.\n",
|
| 703 |
+
"step 1100 | loss 0.8594 | smoothed loss: 1.0289\n",
|
| 704 |
+
"The last 100 steps took: 85.08 seconds.\n",
|
| 705 |
+
"step 1200 | loss 1.4141 | smoothed loss: 1.1086\n",
|
| 706 |
+
"The last 100 steps took: 84.17 seconds.\n",
|
| 707 |
+
"step 1300 | loss 0.8047 | smoothed loss: 1.1398\n",
|
| 708 |
+
"The last 100 steps took: 90.32 seconds.\n",
|
| 709 |
+
"step 1400 | loss 1.7266 | smoothed loss: 1.3914\n",
|
| 710 |
+
"The last 100 steps took: 76.20 seconds.\n",
|
| 711 |
+
"step 1500 | loss 0.8711 | smoothed loss: 1.1715\n",
|
| 712 |
+
"The last 100 steps took: 91.56 seconds.\n",
|
| 713 |
+
"step 1600 | loss 1.1641 | smoothed loss: 1.1652\n",
|
| 714 |
+
"The last 100 steps took: 90.52 seconds.\n",
|
| 715 |
+
"step 1700 | loss 1.3750 | smoothed loss: 1.4680\n",
|
| 716 |
+
"The last 100 steps took: 91.58 seconds.\n",
|
| 717 |
+
"step 1800 | loss 1.2969 | smoothed loss: 1.2055\n",
|
| 718 |
+
"The last 100 steps took: 93.18 seconds.\n",
|
| 719 |
+
"step 1900 | loss 1.3594 | smoothed loss: 1.2840\n",
|
| 720 |
+
"The last 100 steps took: 75.23 seconds.\n",
|
| 721 |
+
"step 2000 | loss 1.1875 | smoothed loss: 1.2492\n",
|
| 722 |
+
"The last 100 steps took: 93.22 seconds.\n",
|
| 723 |
+
"step 2100 | loss 1.5312 | smoothed loss: 1.2500\n",
|
| 724 |
+
"The last 100 steps took: 89.42 seconds.\n",
|
| 725 |
+
"step 2200 | loss 1.2578 | smoothed loss: 1.0488\n",
|
| 726 |
+
"The last 100 steps took: 88.84 seconds.\n",
|
| 727 |
+
"step 2300 | loss 1.3516 | smoothed loss: 1.1641\n",
|
| 728 |
+
"The last 100 steps took: 76.10 seconds.\n",
|
| 729 |
+
"step 2400 | loss 0.7969 | smoothed loss: 1.1305\n",
|
| 730 |
+
"The last 100 steps took: 94.21 seconds.\n",
|
| 731 |
+
"step 2500 | loss 1.4297 | smoothed loss: 1.2496\n",
|
| 732 |
+
"The last 100 steps took: 92.31 seconds.\n",
|
| 733 |
+
"step 2600 | loss 1.1406 | smoothed loss: 1.2168\n",
|
| 734 |
+
"The last 100 steps took: 91.25 seconds.\n",
|
| 735 |
+
"step 2700 | loss 1.2031 | smoothed loss: 1.2645\n",
|
| 736 |
+
"The last 100 steps took: 82.86 seconds.\n",
|
| 737 |
+
"step 2800 | loss 0.8164 | smoothed loss: 1.0625\n",
|
| 738 |
+
"The last 100 steps took: 75.41 seconds.\n",
|
| 739 |
+
"step 2900 | loss 1.0547 | smoothed loss: 1.3023\n",
|
| 740 |
+
"The last 100 steps took: 91.18 seconds.\n",
|
| 741 |
+
"step 3000 | loss 0.7578 | smoothed loss: 1.2715\n",
|
| 742 |
+
"The last 100 steps took: 89.79 seconds.\n",
|
| 743 |
+
"step 3100 | loss 1.0391 | smoothed loss: 1.2273\n",
|
| 744 |
+
"The last 100 steps took: 85.47 seconds.\n",
|
| 745 |
+
"step 3200 | loss 0.9609 | smoothed loss: 1.0578\n",
|
| 746 |
+
"The last 100 steps took: 92.26 seconds.\n",
|
| 747 |
+
"step 3300 | loss 1.5469 | smoothed loss: 1.1762\n",
|
| 748 |
+
"The last 100 steps took: 74.85 seconds.\n",
|
| 749 |
+
"step 3400 | loss 1.8828 | smoothed loss: 1.2742\n",
|
| 750 |
+
"The last 100 steps took: 90.77 seconds.\n",
|
| 751 |
+
"step 3500 | loss 1.6094 | smoothed loss: 1.3133\n",
|
| 752 |
+
"The last 100 steps took: 95.12 seconds.\n",
|
| 753 |
+
"step 3600 | loss 1.2031 | smoothed loss: 1.1922\n",
|
| 754 |
+
"The last 100 steps took: 91.85 seconds.\n",
|
| 755 |
+
"step 3700 | loss 1.6250 | smoothed loss: 1.3082\n",
|
| 756 |
+
"The last 100 steps took: 92.74 seconds.\n",
|
| 757 |
+
"step 3800 | loss 1.4688 | smoothed loss: 1.2098\n",
|
| 758 |
+
"The last 100 steps took: 75.41 seconds.\n",
|
| 759 |
+
"step 3900 | loss 1.2578 | smoothed loss: 1.0668\n",
|
| 760 |
+
"The last 100 steps took: 90.95 seconds.\n",
|
| 761 |
+
"step 4000 | loss 1.1562 | smoothed loss: 1.1832\n",
|
| 762 |
+
"The last 100 steps took: 92.12 seconds.\n",
|
| 763 |
+
"step 4100 | loss 1.2969 | smoothed loss: 1.2406\n",
|
| 764 |
+
"The last 100 steps took: 91.43 seconds.\n",
|
| 765 |
+
"step 4200 | loss 1.5234 | smoothed loss: 1.1359\n",
|
| 766 |
+
"The last 100 steps took: 89.33 seconds.\n",
|
| 767 |
+
"step 4300 | loss 0.9453 | smoothed loss: 1.0145\n",
|
| 768 |
+
"The last 100 steps took: 75.67 seconds.\n",
|
| 769 |
+
"step 4400 | loss 1.6562 | smoothed loss: 1.1062\n",
|
| 770 |
+
"The last 100 steps took: 91.63 seconds.\n",
|
| 771 |
+
"step 4500 | loss 0.7305 | smoothed loss: 1.1695\n",
|
| 772 |
+
"The last 100 steps took: 90.45 seconds.\n",
|
| 773 |
+
"step 4600 | loss 1.3906 | smoothed loss: 1.1898\n",
|
| 774 |
+
"The last 100 steps took: 83.71 seconds.\n",
|
| 775 |
+
"step 4700 | loss 1.1328 | smoothed loss: 1.0383\n",
|
| 776 |
+
"The last 100 steps took: 75.69 seconds.\n",
|
| 777 |
+
"step 4800 | loss 0.6484 | smoothed loss: 1.0418\n",
|
| 778 |
+
"The last 100 steps took: 88.35 seconds.\n",
|
| 779 |
+
"step 4900 | loss 1.5859 | smoothed loss: 1.1891\n",
|
| 780 |
+
"The last 100 steps took: 90.32 seconds.\n",
|
| 781 |
+
"step 5000 | loss 0.8711 | smoothed loss: 1.2063\n",
|
| 782 |
+
"The last 100 steps took: 91.77 seconds.\n"
|
| 783 |
+
]
|
| 784 |
+
}
|
| 785 |
+
],
|
| 786 |
+
"source": [
|
| 787 |
+
"from datasets import load_dataset\n",
|
| 788 |
+
"from transformers import AutoTokenizer\n",
|
| 789 |
+
"import torch.optim as optim\n",
|
| 790 |
+
"import time\n",
|
| 791 |
+
"import json\n",
|
| 792 |
+
"import numpy as np\n",
|
| 793 |
+
"\n",
|
| 794 |
+
"vision_encoder_ckpt = \"google/siglip2-base-patch16-256\"\n",
|
| 795 |
+
"language_model_ckpt = \"HuggingFaceTB/SmolLM2-135M-Instruct\"\n",
|
| 796 |
+
"dataset = load_dataset(\"HuggingFaceM4/FineVisionMax\", split=\"train\", streaming=True)\n",
|
| 797 |
+
"tokenizer = AutoTokenizer.from_pretrained(language_model_ckpt, extra_special_tokens={\"image_token\": \"<|image|>\"})\n",
|
| 798 |
+
"vlm = VisionLanguageModel(vision_encoder_ckpt, language_model_ckpt, tokenizer).to(\"cuda\")\n",
|
| 799 |
+
"optimizer = optim.AdamW(vlm.parameters(), lr=1e-4)\n",
|
| 800 |
+
"\n",
|
| 801 |
+
"def tokenize_sample(sample):\n",
|
| 802 |
+
" messages_list = []\n",
|
| 803 |
+
" for data_dict in sample['texts']:\n",
|
| 804 |
+
" messages_list.append({'role': 'user', 'content': data_dict['user']})\n",
|
| 805 |
+
" messages_list.append({'role': 'assistant', 'content': data_dict['assistant']})\n",
|
| 806 |
+
" messages_list[0]['content'] = tokenizer.image_token*256 + messages_list[0]['content'] # add image tokens\n",
|
| 807 |
+
" tokens = tokenizer.apply_chat_template(messages_list, tokenize=True)\n",
|
| 808 |
+
" return tokens['input_ids']\n",
|
| 809 |
+
"\n",
|
| 810 |
+
"step = 0\n",
|
| 811 |
+
"losses = []\n",
|
| 812 |
+
"max_length = 2048\n",
|
| 813 |
+
"curr_sample = []\n",
|
| 814 |
+
"curr_image = []\n",
|
| 815 |
+
"batch = []\n",
|
| 816 |
+
"image_batch = []\n",
|
| 817 |
+
"max_batch_size = 4\n",
|
| 818 |
+
"batch_complete = False\n",
|
| 819 |
+
"start = time.time()\n",
|
| 820 |
+
"for sample in dataset:\n",
|
| 821 |
+
" if len(sample['images']) != 1:\n",
|
| 822 |
+
" continue\n",
|
| 823 |
+
" tokens = tokenize_sample(sample)\n",
|
| 824 |
+
" if len(tokens) > max_length:\n",
|
| 825 |
+
" continue # Skip samples that are too long\n",
|
| 826 |
+
" if len(curr_sample) + len(tokens) < max_length:\n",
|
| 827 |
+
" curr_sample.extend(tokens)\n",
|
| 828 |
+
" curr_image.extend([sample['images'][0].convert('RGB')])\n",
|
| 829 |
+
" else:\n",
|
| 830 |
+
" batch.append(torch.nn.functional.pad(torch.Tensor(curr_sample), (max_length - len(curr_sample),0), 'constant', tokenizer.pad_token_id))\n",
|
| 831 |
+
" image_batch.extend(curr_image)\n",
|
| 832 |
+
" if len(batch) == max_batch_size:\n",
|
| 833 |
+
" batch_complete = True # We've reached the max batch size, so we can continue\n",
|
| 834 |
+
" curr_sample = tokens\n",
|
| 835 |
+
" curr_image = [sample['images'][0].convert('RGB')]\n",
|
| 836 |
+
" if batch_complete:\n",
|
| 837 |
+
" step += 1\n",
|
| 838 |
+
" if step%100 == 0:\n",
|
| 839 |
+
" print(f\"step {step} | loss {loss.item():.4f} | smoothed loss: {np.mean(losses[-10:]):.4f}\")\n",
|
| 840 |
+
" print(f\"The last 100 steps took: {time.time()-start:.2f} seconds.\")\n",
|
| 841 |
+
" start = time.time()\n",
|
| 842 |
+
"\n",
|
| 843 |
+
" optimizer.zero_grad()\n",
|
| 844 |
+
" logits, loss = vlm(torch.stack(batch).long().cuda(), image_batch)\n",
|
| 845 |
+
" losses.append(loss.item())\n",
|
| 846 |
+
"\n",
|
| 847 |
+
" loss.backward()\n",
|
| 848 |
+
" optimizer.step()\n",
|
| 849 |
+
"\n",
|
| 850 |
+
" batch_complete = False\n",
|
| 851 |
+
" batch = []\n",
|
| 852 |
+
" image_batch = []\n",
|
| 853 |
+
" if step > 5000:\n",
|
| 854 |
+
" break\n",
|
| 855 |
+
"\n",
|
| 856 |
+
"json.dump([round(l, 4) for l in losses], open(\"chapter_3_batched_loss.json\", \"w\"))"
|
| 857 |
+
]
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"cell_type": "code",
|
| 861 |
+
"execution_count": 9,
|
| 862 |
+
"metadata": {},
|
| 863 |
+
"outputs": [],
|
| 864 |
+
"source": [
|
| 865 |
+
"json.dump([round(l, 4) for l in losses], open(\"chapter_3_batched_loss.json\", \"w\"))"
|
| 866 |
+
]
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"cell_type": "code",
|
| 870 |
+
"execution_count": null,
|
| 871 |
+
"metadata": {
|
| 872 |
+
"id": "8uhcUAcSzWo8"
|
| 873 |
+
},
|
| 874 |
+
"outputs": [],
|
| 875 |
+
"source": [
|
| 876 |
+
"from transformers.image_utils import load_image\n",
|
| 877 |
+
"\n",
|
| 878 |
+
"image = load_image(\"https://huggingface.co/datasets/merve/coco/resolve/main/val2017/000000002006.jpg\")\n",
|
| 879 |
+
"\n",
|
| 880 |
+
"def generate(trained_vlm, tokenizer, input_text, image, max_new_tokens):\n",
|
| 881 |
+
" text = [{\"role\": \"user\", \"content\": tokenizer.image_token*256 + input_text}]\n",
|
| 882 |
+
" text_str = tokenizer.apply_chat_template(text, add_generation_prompt=True, tokenize=False)\n",
|
| 883 |
+
" input_tokens = tokenizer(text_str, return_tensors=\"pt\").input_ids.cuda() \n",
|
| 884 |
+
" next_token = None\n",
|
| 885 |
+
" output = []\n",
|
| 886 |
+
" end_of_sentence_token = tokenizer.eos_token_id\n",
|
| 887 |
+
"\n",
|
| 888 |
+
" while len(output) < max_new_tokens:\n",
|
| 889 |
+
" prediction = trained_vlm(input_tokens, image)[0]\n",
|
| 890 |
+
" next_token = torch.argmax(prediction[:, -1, :])\n",
|
| 891 |
+
" if next_token == end_of_sentence_token:\n",
|
| 892 |
+
" break\n",
|
| 893 |
+
" output.append(next_token)\n",
|
| 894 |
+
" input_tokens = torch.cat([input_tokens, next_token[None, None]], dim=1)\n",
|
| 895 |
+
"\n",
|
| 896 |
+
" return ''.join(tokenizer.batch_decode(output))\n",
|
| 897 |
+
"\n",
|
| 898 |
+
"print(generate(vlm, tokenizer, \"What do you see here?\", image, 32))\n"
|
| 899 |
+
]
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"cell_type": "code",
|
| 903 |
+
"execution_count": null,
|
| 904 |
+
"metadata": {},
|
| 905 |
+
"outputs": [],
|
| 906 |
+
"source": []
|
| 907 |
+
}
|
| 908 |
+
],
|
| 909 |
+
"metadata": {
|
| 910 |
+
"accelerator": "GPU",
|
| 911 |
+
"colab": {
|
| 912 |
+
"gpuType": "T4",
|
| 913 |
+
"provenance": []
|
| 914 |
+
},
|
| 915 |
+
"kernelspec": {
|
| 916 |
+
"display_name": "Python 3 (ipykernel)",
|
| 917 |
+
"language": "python",
|
| 918 |
+
"name": "python3"
|
| 919 |
+
},
|
| 920 |
+
"language_info": {
|
| 921 |
+
"codemirror_mode": {
|
| 922 |
+
"name": "ipython",
|
| 923 |
+
"version": 3
|
| 924 |
+
},
|
| 925 |
+
"file_extension": ".py",
|
| 926 |
+
"mimetype": "text/x-python",
|
| 927 |
+
"name": "python",
|
| 928 |
+
"nbconvert_exporter": "python",
|
| 929 |
+
"pygments_lexer": "ipython3",
|
| 930 |
+
"version": "3.10.19"
|
| 931 |
+
},
|
| 932 |
+
"widgets": {
|
| 933 |
+
"application/vnd.jupyter.widget-state+json": {
|
| 934 |
+
"032dcdea26234490a141913d7e361a15": {
|
| 935 |
+
"model_module": "@jupyter-widgets/controls",
|
| 936 |
+
"model_module_version": "1.5.0",
|
| 937 |
+
"model_name": "DescriptionStyleModel",
|
| 938 |
+
"state": {
|
| 939 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 940 |
+
"_model_module_version": "1.5.0",
|
| 941 |
+
"_model_name": "DescriptionStyleModel",
|
| 942 |
+
"_view_count": null,
|
| 943 |
+
"_view_module": "@jupyter-widgets/base",
|
| 944 |
+
"_view_module_version": "1.2.0",
|
| 945 |
+
"_view_name": "StyleView",
|
| 946 |
+
"description_width": ""
|
| 947 |
+
}
|
| 948 |
+
},
|
| 949 |
+
"059fc34bbe9c40a0a81c5640003ccc47": {
|
| 950 |
+
"model_module": "@jupyter-widgets/controls",
|
| 951 |
+
"model_module_version": "1.5.0",
|
| 952 |
+
"model_name": "HBoxModel",
|
| 953 |
+
"state": {
|
| 954 |
+
"_dom_classes": [],
|
| 955 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 956 |
+
"_model_module_version": "1.5.0",
|
| 957 |
+
"_model_name": "HBoxModel",
|
| 958 |
+
"_view_count": null,
|
| 959 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 960 |
+
"_view_module_version": "1.5.0",
|
| 961 |
+
"_view_name": "HBoxView",
|
| 962 |
+
"box_style": "",
|
| 963 |
+
"children": [
|
| 964 |
+
"IPY_MODEL_3cc6112c0037496a9337382c1d16a3d6",
|
| 965 |
+
"IPY_MODEL_f4fcf974faad42a08fb6863763eedb81",
|
| 966 |
+
"IPY_MODEL_d6cc3547f86c4629a8e8e53592167b43"
|
| 967 |
+
],
|
| 968 |
+
"layout": "IPY_MODEL_c04849b1932040188a760a2ebabc91ec"
|
| 969 |
+
}
|
| 970 |
+
},
|
| 971 |
+
"18111ed761e146698786fb4a5d0c7051": {
|
| 972 |
+
"model_module": "@jupyter-widgets/base",
|
| 973 |
+
"model_module_version": "1.2.0",
|
| 974 |
+
"model_name": "LayoutModel",
|
| 975 |
+
"state": {
|
| 976 |
+
"_model_module": "@jupyter-widgets/base",
|
| 977 |
+
"_model_module_version": "1.2.0",
|
| 978 |
+
"_model_name": "LayoutModel",
|
| 979 |
+
"_view_count": null,
|
| 980 |
+
"_view_module": "@jupyter-widgets/base",
|
| 981 |
+
"_view_module_version": "1.2.0",
|
| 982 |
+
"_view_name": "LayoutView",
|
| 983 |
+
"align_content": null,
|
| 984 |
+
"align_items": null,
|
| 985 |
+
"align_self": null,
|
| 986 |
+
"border": null,
|
| 987 |
+
"bottom": null,
|
| 988 |
+
"display": null,
|
| 989 |
+
"flex": null,
|
| 990 |
+
"flex_flow": null,
|
| 991 |
+
"grid_area": null,
|
| 992 |
+
"grid_auto_columns": null,
|
| 993 |
+
"grid_auto_flow": null,
|
| 994 |
+
"grid_auto_rows": null,
|
| 995 |
+
"grid_column": null,
|
| 996 |
+
"grid_gap": null,
|
| 997 |
+
"grid_row": null,
|
| 998 |
+
"grid_template_areas": null,
|
| 999 |
+
"grid_template_columns": null,
|
| 1000 |
+
"grid_template_rows": null,
|
| 1001 |
+
"height": null,
|
| 1002 |
+
"justify_content": null,
|
| 1003 |
+
"justify_items": null,
|
| 1004 |
+
"left": null,
|
| 1005 |
+
"margin": null,
|
| 1006 |
+
"max_height": null,
|
| 1007 |
+
"max_width": null,
|
| 1008 |
+
"min_height": null,
|
| 1009 |
+
"min_width": null,
|
| 1010 |
+
"object_fit": null,
|
| 1011 |
+
"object_position": null,
|
| 1012 |
+
"order": null,
|
| 1013 |
+
"overflow": null,
|
| 1014 |
+
"overflow_x": null,
|
| 1015 |
+
"overflow_y": null,
|
| 1016 |
+
"padding": null,
|
| 1017 |
+
"right": null,
|
| 1018 |
+
"top": null,
|
| 1019 |
+
"visibility": null,
|
| 1020 |
+
"width": null
|
| 1021 |
+
}
|
| 1022 |
+
},
|
| 1023 |
+
"1e523935a83d434c976f10797cbe2286": {
|
| 1024 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1025 |
+
"model_module_version": "1.5.0",
|
| 1026 |
+
"model_name": "ProgressStyleModel",
|
| 1027 |
+
"state": {
|
| 1028 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1029 |
+
"_model_module_version": "1.5.0",
|
| 1030 |
+
"_model_name": "ProgressStyleModel",
|
| 1031 |
+
"_view_count": null,
|
| 1032 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1033 |
+
"_view_module_version": "1.2.0",
|
| 1034 |
+
"_view_name": "StyleView",
|
| 1035 |
+
"bar_color": null,
|
| 1036 |
+
"description_width": ""
|
| 1037 |
+
}
|
| 1038 |
+
},
|
| 1039 |
+
"296ba3ca9f34496ba8edfeee4c76a5d2": {
|
| 1040 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1041 |
+
"model_module_version": "1.5.0",
|
| 1042 |
+
"model_name": "HTMLModel",
|
| 1043 |
+
"state": {
|
| 1044 |
+
"_dom_classes": [],
|
| 1045 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1046 |
+
"_model_module_version": "1.5.0",
|
| 1047 |
+
"_model_name": "HTMLModel",
|
| 1048 |
+
"_view_count": null,
|
| 1049 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1050 |
+
"_view_module_version": "1.5.0",
|
| 1051 |
+
"_view_name": "HTMLView",
|
| 1052 |
+
"description": "",
|
| 1053 |
+
"description_tooltip": null,
|
| 1054 |
+
"layout": "IPY_MODEL_d8ee4430ef8c47bc96d6d7ae2060f9ad",
|
| 1055 |
+
"placeholder": "",
|
| 1056 |
+
"style": "IPY_MODEL_667ff0bc56f94ab582252968d3dca89b",
|
| 1057 |
+
"value": " 272/272 [00:01<00:00, 202.94it/s, Materializing param=model.norm.weight]"
|
| 1058 |
+
}
|
| 1059 |
+
},
|
| 1060 |
+
"2f69578d14bb4ad6967e00931f611f63": {
|
| 1061 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1062 |
+
"model_module_version": "1.5.0",
|
| 1063 |
+
"model_name": "DescriptionStyleModel",
|
| 1064 |
+
"state": {
|
| 1065 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1066 |
+
"_model_module_version": "1.5.0",
|
| 1067 |
+
"_model_name": "DescriptionStyleModel",
|
| 1068 |
+
"_view_count": null,
|
| 1069 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1070 |
+
"_view_module_version": "1.2.0",
|
| 1071 |
+
"_view_name": "StyleView",
|
| 1072 |
+
"description_width": ""
|
| 1073 |
+
}
|
| 1074 |
+
},
|
| 1075 |
+
"3686ac8683c745ee8bc5638d1e962c54": {
|
| 1076 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1077 |
+
"model_module_version": "1.5.0",
|
| 1078 |
+
"model_name": "FloatProgressModel",
|
| 1079 |
+
"state": {
|
| 1080 |
+
"_dom_classes": [],
|
| 1081 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1082 |
+
"_model_module_version": "1.5.0",
|
| 1083 |
+
"_model_name": "FloatProgressModel",
|
| 1084 |
+
"_view_count": null,
|
| 1085 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1086 |
+
"_view_module_version": "1.5.0",
|
| 1087 |
+
"_view_name": "ProgressView",
|
| 1088 |
+
"bar_style": "success",
|
| 1089 |
+
"description": "",
|
| 1090 |
+
"description_tooltip": null,
|
| 1091 |
+
"layout": "IPY_MODEL_b391ef77f8ed4d4f9ccd41ace6a80075",
|
| 1092 |
+
"max": 10000,
|
| 1093 |
+
"min": 0,
|
| 1094 |
+
"orientation": "horizontal",
|
| 1095 |
+
"style": "IPY_MODEL_7f01a3c81bb34621aa7f2744722de6bf",
|
| 1096 |
+
"value": 10000
|
| 1097 |
+
}
|
| 1098 |
+
},
|
| 1099 |
+
"3cc6112c0037496a9337382c1d16a3d6": {
|
| 1100 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1101 |
+
"model_module_version": "1.5.0",
|
| 1102 |
+
"model_name": "HTMLModel",
|
| 1103 |
+
"state": {
|
| 1104 |
+
"_dom_classes": [],
|
| 1105 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1106 |
+
"_model_module_version": "1.5.0",
|
| 1107 |
+
"_model_name": "HTMLModel",
|
| 1108 |
+
"_view_count": null,
|
| 1109 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1110 |
+
"_view_module_version": "1.5.0",
|
| 1111 |
+
"_view_name": "HTMLView",
|
| 1112 |
+
"description": "",
|
| 1113 |
+
"description_tooltip": null,
|
| 1114 |
+
"layout": "IPY_MODEL_5ae3364706d04df0afcd0e941c09a01a",
|
| 1115 |
+
"placeholder": "",
|
| 1116 |
+
"style": "IPY_MODEL_f09cbd8806bc41808b5594cfb16c1a83",
|
| 1117 |
+
"value": "Loading weights: 100%"
|
| 1118 |
+
}
|
| 1119 |
+
},
|
| 1120 |
+
"4d95c6fe1c434425b1368f07952e250a": {
|
| 1121 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1122 |
+
"model_module_version": "1.5.0",
|
| 1123 |
+
"model_name": "DescriptionStyleModel",
|
| 1124 |
+
"state": {
|
| 1125 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1126 |
+
"_model_module_version": "1.5.0",
|
| 1127 |
+
"_model_name": "DescriptionStyleModel",
|
| 1128 |
+
"_view_count": null,
|
| 1129 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1130 |
+
"_view_module_version": "1.2.0",
|
| 1131 |
+
"_view_name": "StyleView",
|
| 1132 |
+
"description_width": ""
|
| 1133 |
+
}
|
| 1134 |
+
},
|
| 1135 |
+
"59c2f3d1128a45ebbc2f1405ae42751a": {
|
| 1136 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1137 |
+
"model_module_version": "1.5.0",
|
| 1138 |
+
"model_name": "ProgressStyleModel",
|
| 1139 |
+
"state": {
|
| 1140 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1141 |
+
"_model_module_version": "1.5.0",
|
| 1142 |
+
"_model_name": "ProgressStyleModel",
|
| 1143 |
+
"_view_count": null,
|
| 1144 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1145 |
+
"_view_module_version": "1.2.0",
|
| 1146 |
+
"_view_name": "StyleView",
|
| 1147 |
+
"bar_color": null,
|
| 1148 |
+
"description_width": ""
|
| 1149 |
+
}
|
| 1150 |
+
},
|
| 1151 |
+
"5ae3364706d04df0afcd0e941c09a01a": {
|
| 1152 |
+
"model_module": "@jupyter-widgets/base",
|
| 1153 |
+
"model_module_version": "1.2.0",
|
| 1154 |
+
"model_name": "LayoutModel",
|
| 1155 |
+
"state": {
|
| 1156 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1157 |
+
"_model_module_version": "1.2.0",
|
| 1158 |
+
"_model_name": "LayoutModel",
|
| 1159 |
+
"_view_count": null,
|
| 1160 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1161 |
+
"_view_module_version": "1.2.0",
|
| 1162 |
+
"_view_name": "LayoutView",
|
| 1163 |
+
"align_content": null,
|
| 1164 |
+
"align_items": null,
|
| 1165 |
+
"align_self": null,
|
| 1166 |
+
"border": null,
|
| 1167 |
+
"bottom": null,
|
| 1168 |
+
"display": null,
|
| 1169 |
+
"flex": null,
|
| 1170 |
+
"flex_flow": null,
|
| 1171 |
+
"grid_area": null,
|
| 1172 |
+
"grid_auto_columns": null,
|
| 1173 |
+
"grid_auto_flow": null,
|
| 1174 |
+
"grid_auto_rows": null,
|
| 1175 |
+
"grid_column": null,
|
| 1176 |
+
"grid_gap": null,
|
| 1177 |
+
"grid_row": null,
|
| 1178 |
+
"grid_template_areas": null,
|
| 1179 |
+
"grid_template_columns": null,
|
| 1180 |
+
"grid_template_rows": null,
|
| 1181 |
+
"height": null,
|
| 1182 |
+
"justify_content": null,
|
| 1183 |
+
"justify_items": null,
|
| 1184 |
+
"left": null,
|
| 1185 |
+
"margin": null,
|
| 1186 |
+
"max_height": null,
|
| 1187 |
+
"max_width": null,
|
| 1188 |
+
"min_height": null,
|
| 1189 |
+
"min_width": null,
|
| 1190 |
+
"object_fit": null,
|
| 1191 |
+
"object_position": null,
|
| 1192 |
+
"order": null,
|
| 1193 |
+
"overflow": null,
|
| 1194 |
+
"overflow_x": null,
|
| 1195 |
+
"overflow_y": null,
|
| 1196 |
+
"padding": null,
|
| 1197 |
+
"right": null,
|
| 1198 |
+
"top": null,
|
| 1199 |
+
"visibility": null,
|
| 1200 |
+
"width": null
|
| 1201 |
+
}
|
| 1202 |
+
},
|
| 1203 |
+
"5cb1c4db303b4b6aaaabc7cf88e975d5": {
|
| 1204 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1205 |
+
"model_module_version": "1.5.0",
|
| 1206 |
+
"model_name": "DescriptionStyleModel",
|
| 1207 |
+
"state": {
|
| 1208 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1209 |
+
"_model_module_version": "1.5.0",
|
| 1210 |
+
"_model_name": "DescriptionStyleModel",
|
| 1211 |
+
"_view_count": null,
|
| 1212 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1213 |
+
"_view_module_version": "1.2.0",
|
| 1214 |
+
"_view_name": "StyleView",
|
| 1215 |
+
"description_width": ""
|
| 1216 |
+
}
|
| 1217 |
+
},
|
| 1218 |
+
"64363326d62641f3aee10f317d92140e": {
|
| 1219 |
+
"model_module": "@jupyter-widgets/base",
|
| 1220 |
+
"model_module_version": "1.2.0",
|
| 1221 |
+
"model_name": "LayoutModel",
|
| 1222 |
+
"state": {
|
| 1223 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1224 |
+
"_model_module_version": "1.2.0",
|
| 1225 |
+
"_model_name": "LayoutModel",
|
| 1226 |
+
"_view_count": null,
|
| 1227 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1228 |
+
"_view_module_version": "1.2.0",
|
| 1229 |
+
"_view_name": "LayoutView",
|
| 1230 |
+
"align_content": null,
|
| 1231 |
+
"align_items": null,
|
| 1232 |
+
"align_self": null,
|
| 1233 |
+
"border": null,
|
| 1234 |
+
"bottom": null,
|
| 1235 |
+
"display": null,
|
| 1236 |
+
"flex": null,
|
| 1237 |
+
"flex_flow": null,
|
| 1238 |
+
"grid_area": null,
|
| 1239 |
+
"grid_auto_columns": null,
|
| 1240 |
+
"grid_auto_flow": null,
|
| 1241 |
+
"grid_auto_rows": null,
|
| 1242 |
+
"grid_column": null,
|
| 1243 |
+
"grid_gap": null,
|
| 1244 |
+
"grid_row": null,
|
| 1245 |
+
"grid_template_areas": null,
|
| 1246 |
+
"grid_template_columns": null,
|
| 1247 |
+
"grid_template_rows": null,
|
| 1248 |
+
"height": null,
|
| 1249 |
+
"justify_content": null,
|
| 1250 |
+
"justify_items": null,
|
| 1251 |
+
"left": null,
|
| 1252 |
+
"margin": null,
|
| 1253 |
+
"max_height": null,
|
| 1254 |
+
"max_width": null,
|
| 1255 |
+
"min_height": null,
|
| 1256 |
+
"min_width": null,
|
| 1257 |
+
"object_fit": null,
|
| 1258 |
+
"object_position": null,
|
| 1259 |
+
"order": null,
|
| 1260 |
+
"overflow": null,
|
| 1261 |
+
"overflow_x": null,
|
| 1262 |
+
"overflow_y": null,
|
| 1263 |
+
"padding": null,
|
| 1264 |
+
"right": null,
|
| 1265 |
+
"top": null,
|
| 1266 |
+
"visibility": null,
|
| 1267 |
+
"width": null
|
| 1268 |
+
}
|
| 1269 |
+
},
|
| 1270 |
+
"667ff0bc56f94ab582252968d3dca89b": {
|
| 1271 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1272 |
+
"model_module_version": "1.5.0",
|
| 1273 |
+
"model_name": "DescriptionStyleModel",
|
| 1274 |
+
"state": {
|
| 1275 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1276 |
+
"_model_module_version": "1.5.0",
|
| 1277 |
+
"_model_name": "DescriptionStyleModel",
|
| 1278 |
+
"_view_count": null,
|
| 1279 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1280 |
+
"_view_module_version": "1.2.0",
|
| 1281 |
+
"_view_name": "StyleView",
|
| 1282 |
+
"description_width": ""
|
| 1283 |
+
}
|
| 1284 |
+
},
|
| 1285 |
+
"6af6b0d5fb5942f0874062514fecdac9": {
|
| 1286 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1287 |
+
"model_module_version": "1.5.0",
|
| 1288 |
+
"model_name": "HBoxModel",
|
| 1289 |
+
"state": {
|
| 1290 |
+
"_dom_classes": [],
|
| 1291 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1292 |
+
"_model_module_version": "1.5.0",
|
| 1293 |
+
"_model_name": "HBoxModel",
|
| 1294 |
+
"_view_count": null,
|
| 1295 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1296 |
+
"_view_module_version": "1.5.0",
|
| 1297 |
+
"_view_name": "HBoxView",
|
| 1298 |
+
"box_style": "",
|
| 1299 |
+
"children": [
|
| 1300 |
+
"IPY_MODEL_aaceb8bb8d144e62a7fc19dde30e981b",
|
| 1301 |
+
"IPY_MODEL_9a9026e81b974c7ba30f92d2e7d5f16f",
|
| 1302 |
+
"IPY_MODEL_81bebd074665482b9af8e6d39b6b171a"
|
| 1303 |
+
],
|
| 1304 |
+
"layout": "IPY_MODEL_c1891461614e45189d47fa3948da6808"
|
| 1305 |
+
}
|
| 1306 |
+
},
|
| 1307 |
+
"6baf2c3128d1425c8f979ed703b0fc6b": {
|
| 1308 |
+
"model_module": "@jupyter-widgets/base",
|
| 1309 |
+
"model_module_version": "1.2.0",
|
| 1310 |
+
"model_name": "LayoutModel",
|
| 1311 |
+
"state": {
|
| 1312 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1313 |
+
"_model_module_version": "1.2.0",
|
| 1314 |
+
"_model_name": "LayoutModel",
|
| 1315 |
+
"_view_count": null,
|
| 1316 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1317 |
+
"_view_module_version": "1.2.0",
|
| 1318 |
+
"_view_name": "LayoutView",
|
| 1319 |
+
"align_content": null,
|
| 1320 |
+
"align_items": null,
|
| 1321 |
+
"align_self": null,
|
| 1322 |
+
"border": null,
|
| 1323 |
+
"bottom": null,
|
| 1324 |
+
"display": null,
|
| 1325 |
+
"flex": null,
|
| 1326 |
+
"flex_flow": null,
|
| 1327 |
+
"grid_area": null,
|
| 1328 |
+
"grid_auto_columns": null,
|
| 1329 |
+
"grid_auto_flow": null,
|
| 1330 |
+
"grid_auto_rows": null,
|
| 1331 |
+
"grid_column": null,
|
| 1332 |
+
"grid_gap": null,
|
| 1333 |
+
"grid_row": null,
|
| 1334 |
+
"grid_template_areas": null,
|
| 1335 |
+
"grid_template_columns": null,
|
| 1336 |
+
"grid_template_rows": null,
|
| 1337 |
+
"height": null,
|
| 1338 |
+
"justify_content": null,
|
| 1339 |
+
"justify_items": null,
|
| 1340 |
+
"left": null,
|
| 1341 |
+
"margin": null,
|
| 1342 |
+
"max_height": null,
|
| 1343 |
+
"max_width": null,
|
| 1344 |
+
"min_height": null,
|
| 1345 |
+
"min_width": null,
|
| 1346 |
+
"object_fit": null,
|
| 1347 |
+
"object_position": null,
|
| 1348 |
+
"order": null,
|
| 1349 |
+
"overflow": null,
|
| 1350 |
+
"overflow_x": null,
|
| 1351 |
+
"overflow_y": null,
|
| 1352 |
+
"padding": null,
|
| 1353 |
+
"right": null,
|
| 1354 |
+
"top": null,
|
| 1355 |
+
"visibility": null,
|
| 1356 |
+
"width": null
|
| 1357 |
+
}
|
| 1358 |
+
},
|
| 1359 |
+
"6d97eea8e68948698db477acaa7ff4f6": {
|
| 1360 |
+
"model_module": "@jupyter-widgets/base",
|
| 1361 |
+
"model_module_version": "1.2.0",
|
| 1362 |
+
"model_name": "LayoutModel",
|
| 1363 |
+
"state": {
|
| 1364 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1365 |
+
"_model_module_version": "1.2.0",
|
| 1366 |
+
"_model_name": "LayoutModel",
|
| 1367 |
+
"_view_count": null,
|
| 1368 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1369 |
+
"_view_module_version": "1.2.0",
|
| 1370 |
+
"_view_name": "LayoutView",
|
| 1371 |
+
"align_content": null,
|
| 1372 |
+
"align_items": null,
|
| 1373 |
+
"align_self": null,
|
| 1374 |
+
"border": null,
|
| 1375 |
+
"bottom": null,
|
| 1376 |
+
"display": null,
|
| 1377 |
+
"flex": null,
|
| 1378 |
+
"flex_flow": null,
|
| 1379 |
+
"grid_area": null,
|
| 1380 |
+
"grid_auto_columns": null,
|
| 1381 |
+
"grid_auto_flow": null,
|
| 1382 |
+
"grid_auto_rows": null,
|
| 1383 |
+
"grid_column": null,
|
| 1384 |
+
"grid_gap": null,
|
| 1385 |
+
"grid_row": null,
|
| 1386 |
+
"grid_template_areas": null,
|
| 1387 |
+
"grid_template_columns": null,
|
| 1388 |
+
"grid_template_rows": null,
|
| 1389 |
+
"height": null,
|
| 1390 |
+
"justify_content": null,
|
| 1391 |
+
"justify_items": null,
|
| 1392 |
+
"left": null,
|
| 1393 |
+
"margin": null,
|
| 1394 |
+
"max_height": null,
|
| 1395 |
+
"max_width": null,
|
| 1396 |
+
"min_height": null,
|
| 1397 |
+
"min_width": null,
|
| 1398 |
+
"object_fit": null,
|
| 1399 |
+
"object_position": null,
|
| 1400 |
+
"order": null,
|
| 1401 |
+
"overflow": null,
|
| 1402 |
+
"overflow_x": null,
|
| 1403 |
+
"overflow_y": null,
|
| 1404 |
+
"padding": null,
|
| 1405 |
+
"right": null,
|
| 1406 |
+
"top": null,
|
| 1407 |
+
"visibility": null,
|
| 1408 |
+
"width": null
|
| 1409 |
+
}
|
| 1410 |
+
},
|
| 1411 |
+
"7f01a3c81bb34621aa7f2744722de6bf": {
|
| 1412 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1413 |
+
"model_module_version": "1.5.0",
|
| 1414 |
+
"model_name": "ProgressStyleModel",
|
| 1415 |
+
"state": {
|
| 1416 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1417 |
+
"_model_module_version": "1.5.0",
|
| 1418 |
+
"_model_name": "ProgressStyleModel",
|
| 1419 |
+
"_view_count": null,
|
| 1420 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1421 |
+
"_view_module_version": "1.2.0",
|
| 1422 |
+
"_view_name": "StyleView",
|
| 1423 |
+
"bar_color": null,
|
| 1424 |
+
"description_width": ""
|
| 1425 |
+
}
|
| 1426 |
+
},
|
| 1427 |
+
"81bebd074665482b9af8e6d39b6b171a": {
|
| 1428 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1429 |
+
"model_module_version": "1.5.0",
|
| 1430 |
+
"model_name": "HTMLModel",
|
| 1431 |
+
"state": {
|
| 1432 |
+
"_dom_classes": [],
|
| 1433 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1434 |
+
"_model_module_version": "1.5.0",
|
| 1435 |
+
"_model_name": "HTMLModel",
|
| 1436 |
+
"_view_count": null,
|
| 1437 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1438 |
+
"_view_module_version": "1.5.0",
|
| 1439 |
+
"_view_name": "HTMLView",
|
| 1440 |
+
"description": "",
|
| 1441 |
+
"description_tooltip": null,
|
| 1442 |
+
"layout": "IPY_MODEL_8de705db750048ff890e74afcf97f964",
|
| 1443 |
+
"placeholder": "",
|
| 1444 |
+
"style": "IPY_MODEL_032dcdea26234490a141913d7e361a15",
|
| 1445 |
+
"value": " 4.25k/? [00:00<00:00, 145kB/s]"
|
| 1446 |
+
}
|
| 1447 |
+
},
|
| 1448 |
+
"891d641ad7634c9aa5186589f2aba958": {
|
| 1449 |
+
"model_module": "@jupyter-widgets/base",
|
| 1450 |
+
"model_module_version": "1.2.0",
|
| 1451 |
+
"model_name": "LayoutModel",
|
| 1452 |
+
"state": {
|
| 1453 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1454 |
+
"_model_module_version": "1.2.0",
|
| 1455 |
+
"_model_name": "LayoutModel",
|
| 1456 |
+
"_view_count": null,
|
| 1457 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1458 |
+
"_view_module_version": "1.2.0",
|
| 1459 |
+
"_view_name": "LayoutView",
|
| 1460 |
+
"align_content": null,
|
| 1461 |
+
"align_items": null,
|
| 1462 |
+
"align_self": null,
|
| 1463 |
+
"border": null,
|
| 1464 |
+
"bottom": null,
|
| 1465 |
+
"display": null,
|
| 1466 |
+
"flex": null,
|
| 1467 |
+
"flex_flow": null,
|
| 1468 |
+
"grid_area": null,
|
| 1469 |
+
"grid_auto_columns": null,
|
| 1470 |
+
"grid_auto_flow": null,
|
| 1471 |
+
"grid_auto_rows": null,
|
| 1472 |
+
"grid_column": null,
|
| 1473 |
+
"grid_gap": null,
|
| 1474 |
+
"grid_row": null,
|
| 1475 |
+
"grid_template_areas": null,
|
| 1476 |
+
"grid_template_columns": null,
|
| 1477 |
+
"grid_template_rows": null,
|
| 1478 |
+
"height": null,
|
| 1479 |
+
"justify_content": null,
|
| 1480 |
+
"justify_items": null,
|
| 1481 |
+
"left": null,
|
| 1482 |
+
"margin": null,
|
| 1483 |
+
"max_height": null,
|
| 1484 |
+
"max_width": null,
|
| 1485 |
+
"min_height": null,
|
| 1486 |
+
"min_width": null,
|
| 1487 |
+
"object_fit": null,
|
| 1488 |
+
"object_position": null,
|
| 1489 |
+
"order": null,
|
| 1490 |
+
"overflow": null,
|
| 1491 |
+
"overflow_x": null,
|
| 1492 |
+
"overflow_y": null,
|
| 1493 |
+
"padding": null,
|
| 1494 |
+
"right": null,
|
| 1495 |
+
"top": null,
|
| 1496 |
+
"visibility": null,
|
| 1497 |
+
"width": null
|
| 1498 |
+
}
|
| 1499 |
+
},
|
| 1500 |
+
"8d851590df7c4cf3a9fc293d03befa40": {
|
| 1501 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1502 |
+
"model_module_version": "1.5.0",
|
| 1503 |
+
"model_name": "HBoxModel",
|
| 1504 |
+
"state": {
|
| 1505 |
+
"_dom_classes": [],
|
| 1506 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1507 |
+
"_model_module_version": "1.5.0",
|
| 1508 |
+
"_model_name": "HBoxModel",
|
| 1509 |
+
"_view_count": null,
|
| 1510 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1511 |
+
"_view_module_version": "1.5.0",
|
| 1512 |
+
"_view_name": "HBoxView",
|
| 1513 |
+
"box_style": "",
|
| 1514 |
+
"children": [
|
| 1515 |
+
"IPY_MODEL_aa9c7dbcf7b04f339fce5bee0d24c98d",
|
| 1516 |
+
"IPY_MODEL_fba6cbf72ad54e748031d11669b3e31a",
|
| 1517 |
+
"IPY_MODEL_296ba3ca9f34496ba8edfeee4c76a5d2"
|
| 1518 |
+
],
|
| 1519 |
+
"layout": "IPY_MODEL_cc1e631bd9154962bcb42eb6fc234303"
|
| 1520 |
+
}
|
| 1521 |
+
},
|
| 1522 |
+
"8de705db750048ff890e74afcf97f964": {
|
| 1523 |
+
"model_module": "@jupyter-widgets/base",
|
| 1524 |
+
"model_module_version": "1.2.0",
|
| 1525 |
+
"model_name": "LayoutModel",
|
| 1526 |
+
"state": {
|
| 1527 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1528 |
+
"_model_module_version": "1.2.0",
|
| 1529 |
+
"_model_name": "LayoutModel",
|
| 1530 |
+
"_view_count": null,
|
| 1531 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1532 |
+
"_view_module_version": "1.2.0",
|
| 1533 |
+
"_view_name": "LayoutView",
|
| 1534 |
+
"align_content": null,
|
| 1535 |
+
"align_items": null,
|
| 1536 |
+
"align_self": null,
|
| 1537 |
+
"border": null,
|
| 1538 |
+
"bottom": null,
|
| 1539 |
+
"display": null,
|
| 1540 |
+
"flex": null,
|
| 1541 |
+
"flex_flow": null,
|
| 1542 |
+
"grid_area": null,
|
| 1543 |
+
"grid_auto_columns": null,
|
| 1544 |
+
"grid_auto_flow": null,
|
| 1545 |
+
"grid_auto_rows": null,
|
| 1546 |
+
"grid_column": null,
|
| 1547 |
+
"grid_gap": null,
|
| 1548 |
+
"grid_row": null,
|
| 1549 |
+
"grid_template_areas": null,
|
| 1550 |
+
"grid_template_columns": null,
|
| 1551 |
+
"grid_template_rows": null,
|
| 1552 |
+
"height": null,
|
| 1553 |
+
"justify_content": null,
|
| 1554 |
+
"justify_items": null,
|
| 1555 |
+
"left": null,
|
| 1556 |
+
"margin": null,
|
| 1557 |
+
"max_height": null,
|
| 1558 |
+
"max_width": null,
|
| 1559 |
+
"min_height": null,
|
| 1560 |
+
"min_width": null,
|
| 1561 |
+
"object_fit": null,
|
| 1562 |
+
"object_position": null,
|
| 1563 |
+
"order": null,
|
| 1564 |
+
"overflow": null,
|
| 1565 |
+
"overflow_x": null,
|
| 1566 |
+
"overflow_y": null,
|
| 1567 |
+
"padding": null,
|
| 1568 |
+
"right": null,
|
| 1569 |
+
"top": null,
|
| 1570 |
+
"visibility": null,
|
| 1571 |
+
"width": null
|
| 1572 |
+
}
|
| 1573 |
+
},
|
| 1574 |
+
"9049575e797c4bffa86129fd9394f8fd": {
|
| 1575 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1576 |
+
"model_module_version": "1.5.0",
|
| 1577 |
+
"model_name": "HTMLModel",
|
| 1578 |
+
"state": {
|
| 1579 |
+
"_dom_classes": [],
|
| 1580 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1581 |
+
"_model_module_version": "1.5.0",
|
| 1582 |
+
"_model_name": "HTMLModel",
|
| 1583 |
+
"_view_count": null,
|
| 1584 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1585 |
+
"_view_module_version": "1.5.0",
|
| 1586 |
+
"_view_name": "HTMLView",
|
| 1587 |
+
"description": "",
|
| 1588 |
+
"description_tooltip": null,
|
| 1589 |
+
"layout": "IPY_MODEL_6d97eea8e68948698db477acaa7ff4f6",
|
| 1590 |
+
"placeholder": "",
|
| 1591 |
+
"style": "IPY_MODEL_dbbd386fb13b4dbcae40301c1bbbbf90",
|
| 1592 |
+
"value": " 10000/10000 [00:03<00:00, 3.51s/it]"
|
| 1593 |
+
}
|
| 1594 |
+
},
|
| 1595 |
+
"9a9026e81b974c7ba30f92d2e7d5f16f": {
|
| 1596 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1597 |
+
"model_module_version": "1.5.0",
|
| 1598 |
+
"model_name": "FloatProgressModel",
|
| 1599 |
+
"state": {
|
| 1600 |
+
"_dom_classes": [],
|
| 1601 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1602 |
+
"_model_module_version": "1.5.0",
|
| 1603 |
+
"_model_name": "FloatProgressModel",
|
| 1604 |
+
"_view_count": null,
|
| 1605 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1606 |
+
"_view_module_version": "1.5.0",
|
| 1607 |
+
"_view_name": "ProgressView",
|
| 1608 |
+
"bar_style": "success",
|
| 1609 |
+
"description": "",
|
| 1610 |
+
"description_tooltip": null,
|
| 1611 |
+
"layout": "IPY_MODEL_edcb5218b9bc4e9d9da375eb93de2829",
|
| 1612 |
+
"max": 1,
|
| 1613 |
+
"min": 0,
|
| 1614 |
+
"orientation": "horizontal",
|
| 1615 |
+
"style": "IPY_MODEL_59c2f3d1128a45ebbc2f1405ae42751a",
|
| 1616 |
+
"value": 1
|
| 1617 |
+
}
|
| 1618 |
+
},
|
| 1619 |
+
"9db3a6aae167458b8e3c6b785c25f225": {
|
| 1620 |
+
"model_module": "@jupyter-widgets/base",
|
| 1621 |
+
"model_module_version": "1.2.0",
|
| 1622 |
+
"model_name": "LayoutModel",
|
| 1623 |
+
"state": {
|
| 1624 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1625 |
+
"_model_module_version": "1.2.0",
|
| 1626 |
+
"_model_name": "LayoutModel",
|
| 1627 |
+
"_view_count": null,
|
| 1628 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1629 |
+
"_view_module_version": "1.2.0",
|
| 1630 |
+
"_view_name": "LayoutView",
|
| 1631 |
+
"align_content": null,
|
| 1632 |
+
"align_items": null,
|
| 1633 |
+
"align_self": null,
|
| 1634 |
+
"border": null,
|
| 1635 |
+
"bottom": null,
|
| 1636 |
+
"display": null,
|
| 1637 |
+
"flex": null,
|
| 1638 |
+
"flex_flow": null,
|
| 1639 |
+
"grid_area": null,
|
| 1640 |
+
"grid_auto_columns": null,
|
| 1641 |
+
"grid_auto_flow": null,
|
| 1642 |
+
"grid_auto_rows": null,
|
| 1643 |
+
"grid_column": null,
|
| 1644 |
+
"grid_gap": null,
|
| 1645 |
+
"grid_row": null,
|
| 1646 |
+
"grid_template_areas": null,
|
| 1647 |
+
"grid_template_columns": null,
|
| 1648 |
+
"grid_template_rows": null,
|
| 1649 |
+
"height": null,
|
| 1650 |
+
"justify_content": null,
|
| 1651 |
+
"justify_items": null,
|
| 1652 |
+
"left": null,
|
| 1653 |
+
"margin": null,
|
| 1654 |
+
"max_height": null,
|
| 1655 |
+
"max_width": null,
|
| 1656 |
+
"min_height": null,
|
| 1657 |
+
"min_width": null,
|
| 1658 |
+
"object_fit": null,
|
| 1659 |
+
"object_position": null,
|
| 1660 |
+
"order": null,
|
| 1661 |
+
"overflow": null,
|
| 1662 |
+
"overflow_x": null,
|
| 1663 |
+
"overflow_y": null,
|
| 1664 |
+
"padding": null,
|
| 1665 |
+
"right": null,
|
| 1666 |
+
"top": null,
|
| 1667 |
+
"visibility": null,
|
| 1668 |
+
"width": null
|
| 1669 |
+
}
|
| 1670 |
+
},
|
| 1671 |
+
"a65cbec1df084615915186797051a9e3": {
|
| 1672 |
+
"model_module": "@jupyter-widgets/base",
|
| 1673 |
+
"model_module_version": "1.2.0",
|
| 1674 |
+
"model_name": "LayoutModel",
|
| 1675 |
+
"state": {
|
| 1676 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1677 |
+
"_model_module_version": "1.2.0",
|
| 1678 |
+
"_model_name": "LayoutModel",
|
| 1679 |
+
"_view_count": null,
|
| 1680 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1681 |
+
"_view_module_version": "1.2.0",
|
| 1682 |
+
"_view_name": "LayoutView",
|
| 1683 |
+
"align_content": null,
|
| 1684 |
+
"align_items": null,
|
| 1685 |
+
"align_self": null,
|
| 1686 |
+
"border": null,
|
| 1687 |
+
"bottom": null,
|
| 1688 |
+
"display": null,
|
| 1689 |
+
"flex": null,
|
| 1690 |
+
"flex_flow": null,
|
| 1691 |
+
"grid_area": null,
|
| 1692 |
+
"grid_auto_columns": null,
|
| 1693 |
+
"grid_auto_flow": null,
|
| 1694 |
+
"grid_auto_rows": null,
|
| 1695 |
+
"grid_column": null,
|
| 1696 |
+
"grid_gap": null,
|
| 1697 |
+
"grid_row": null,
|
| 1698 |
+
"grid_template_areas": null,
|
| 1699 |
+
"grid_template_columns": null,
|
| 1700 |
+
"grid_template_rows": null,
|
| 1701 |
+
"height": null,
|
| 1702 |
+
"justify_content": null,
|
| 1703 |
+
"justify_items": null,
|
| 1704 |
+
"left": null,
|
| 1705 |
+
"margin": null,
|
| 1706 |
+
"max_height": null,
|
| 1707 |
+
"max_width": null,
|
| 1708 |
+
"min_height": null,
|
| 1709 |
+
"min_width": null,
|
| 1710 |
+
"object_fit": null,
|
| 1711 |
+
"object_position": null,
|
| 1712 |
+
"order": null,
|
| 1713 |
+
"overflow": null,
|
| 1714 |
+
"overflow_x": null,
|
| 1715 |
+
"overflow_y": null,
|
| 1716 |
+
"padding": null,
|
| 1717 |
+
"right": null,
|
| 1718 |
+
"top": null,
|
| 1719 |
+
"visibility": null,
|
| 1720 |
+
"width": null
|
| 1721 |
+
}
|
| 1722 |
+
},
|
| 1723 |
+
"aa902a918c79418ea4b4516fb5673fa4": {
|
| 1724 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1725 |
+
"model_module_version": "1.5.0",
|
| 1726 |
+
"model_name": "ProgressStyleModel",
|
| 1727 |
+
"state": {
|
| 1728 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1729 |
+
"_model_module_version": "1.5.0",
|
| 1730 |
+
"_model_name": "ProgressStyleModel",
|
| 1731 |
+
"_view_count": null,
|
| 1732 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1733 |
+
"_view_module_version": "1.2.0",
|
| 1734 |
+
"_view_name": "StyleView",
|
| 1735 |
+
"bar_color": null,
|
| 1736 |
+
"description_width": ""
|
| 1737 |
+
}
|
| 1738 |
+
},
|
| 1739 |
+
"aa9c7dbcf7b04f339fce5bee0d24c98d": {
|
| 1740 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1741 |
+
"model_module_version": "1.5.0",
|
| 1742 |
+
"model_name": "HTMLModel",
|
| 1743 |
+
"state": {
|
| 1744 |
+
"_dom_classes": [],
|
| 1745 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1746 |
+
"_model_module_version": "1.5.0",
|
| 1747 |
+
"_model_name": "HTMLModel",
|
| 1748 |
+
"_view_count": null,
|
| 1749 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1750 |
+
"_view_module_version": "1.5.0",
|
| 1751 |
+
"_view_name": "HTMLView",
|
| 1752 |
+
"description": "",
|
| 1753 |
+
"description_tooltip": null,
|
| 1754 |
+
"layout": "IPY_MODEL_a65cbec1df084615915186797051a9e3",
|
| 1755 |
+
"placeholder": "",
|
| 1756 |
+
"style": "IPY_MODEL_4d95c6fe1c434425b1368f07952e250a",
|
| 1757 |
+
"value": "Loading weights: 100%"
|
| 1758 |
+
}
|
| 1759 |
+
},
|
| 1760 |
+
"aaceb8bb8d144e62a7fc19dde30e981b": {
|
| 1761 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1762 |
+
"model_module_version": "1.5.0",
|
| 1763 |
+
"model_name": "HTMLModel",
|
| 1764 |
+
"state": {
|
| 1765 |
+
"_dom_classes": [],
|
| 1766 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1767 |
+
"_model_module_version": "1.5.0",
|
| 1768 |
+
"_model_name": "HTMLModel",
|
| 1769 |
+
"_view_count": null,
|
| 1770 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 1771 |
+
"_view_module_version": "1.5.0",
|
| 1772 |
+
"_view_name": "HTMLView",
|
| 1773 |
+
"description": "",
|
| 1774 |
+
"description_tooltip": null,
|
| 1775 |
+
"layout": "IPY_MODEL_db773254517a428684eb679771212f2c",
|
| 1776 |
+
"placeholder": "",
|
| 1777 |
+
"style": "IPY_MODEL_5cb1c4db303b4b6aaaabc7cf88e975d5",
|
| 1778 |
+
"value": "README.md: "
|
| 1779 |
+
}
|
| 1780 |
+
},
|
| 1781 |
+
"b391ef77f8ed4d4f9ccd41ace6a80075": {
|
| 1782 |
+
"model_module": "@jupyter-widgets/base",
|
| 1783 |
+
"model_module_version": "1.2.0",
|
| 1784 |
+
"model_name": "LayoutModel",
|
| 1785 |
+
"state": {
|
| 1786 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1787 |
+
"_model_module_version": "1.2.0",
|
| 1788 |
+
"_model_name": "LayoutModel",
|
| 1789 |
+
"_view_count": null,
|
| 1790 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1791 |
+
"_view_module_version": "1.2.0",
|
| 1792 |
+
"_view_name": "LayoutView",
|
| 1793 |
+
"align_content": null,
|
| 1794 |
+
"align_items": null,
|
| 1795 |
+
"align_self": null,
|
| 1796 |
+
"border": null,
|
| 1797 |
+
"bottom": null,
|
| 1798 |
+
"display": null,
|
| 1799 |
+
"flex": null,
|
| 1800 |
+
"flex_flow": null,
|
| 1801 |
+
"grid_area": null,
|
| 1802 |
+
"grid_auto_columns": null,
|
| 1803 |
+
"grid_auto_flow": null,
|
| 1804 |
+
"grid_auto_rows": null,
|
| 1805 |
+
"grid_column": null,
|
| 1806 |
+
"grid_gap": null,
|
| 1807 |
+
"grid_row": null,
|
| 1808 |
+
"grid_template_areas": null,
|
| 1809 |
+
"grid_template_columns": null,
|
| 1810 |
+
"grid_template_rows": null,
|
| 1811 |
+
"height": null,
|
| 1812 |
+
"justify_content": null,
|
| 1813 |
+
"justify_items": null,
|
| 1814 |
+
"left": null,
|
| 1815 |
+
"margin": null,
|
| 1816 |
+
"max_height": null,
|
| 1817 |
+
"max_width": null,
|
| 1818 |
+
"min_height": null,
|
| 1819 |
+
"min_width": null,
|
| 1820 |
+
"object_fit": null,
|
| 1821 |
+
"object_position": null,
|
| 1822 |
+
"order": null,
|
| 1823 |
+
"overflow": null,
|
| 1824 |
+
"overflow_x": null,
|
| 1825 |
+
"overflow_y": null,
|
| 1826 |
+
"padding": null,
|
| 1827 |
+
"right": null,
|
| 1828 |
+
"top": null,
|
| 1829 |
+
"visibility": null,
|
| 1830 |
+
"width": null
|
| 1831 |
+
}
|
| 1832 |
+
},
|
| 1833 |
+
"c04849b1932040188a760a2ebabc91ec": {
|
| 1834 |
+
"model_module": "@jupyter-widgets/base",
|
| 1835 |
+
"model_module_version": "1.2.0",
|
| 1836 |
+
"model_name": "LayoutModel",
|
| 1837 |
+
"state": {
|
| 1838 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1839 |
+
"_model_module_version": "1.2.0",
|
| 1840 |
+
"_model_name": "LayoutModel",
|
| 1841 |
+
"_view_count": null,
|
| 1842 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1843 |
+
"_view_module_version": "1.2.0",
|
| 1844 |
+
"_view_name": "LayoutView",
|
| 1845 |
+
"align_content": null,
|
| 1846 |
+
"align_items": null,
|
| 1847 |
+
"align_self": null,
|
| 1848 |
+
"border": null,
|
| 1849 |
+
"bottom": null,
|
| 1850 |
+
"display": null,
|
| 1851 |
+
"flex": null,
|
| 1852 |
+
"flex_flow": null,
|
| 1853 |
+
"grid_area": null,
|
| 1854 |
+
"grid_auto_columns": null,
|
| 1855 |
+
"grid_auto_flow": null,
|
| 1856 |
+
"grid_auto_rows": null,
|
| 1857 |
+
"grid_column": null,
|
| 1858 |
+
"grid_gap": null,
|
| 1859 |
+
"grid_row": null,
|
| 1860 |
+
"grid_template_areas": null,
|
| 1861 |
+
"grid_template_columns": null,
|
| 1862 |
+
"grid_template_rows": null,
|
| 1863 |
+
"height": null,
|
| 1864 |
+
"justify_content": null,
|
| 1865 |
+
"justify_items": null,
|
| 1866 |
+
"left": null,
|
| 1867 |
+
"margin": null,
|
| 1868 |
+
"max_height": null,
|
| 1869 |
+
"max_width": null,
|
| 1870 |
+
"min_height": null,
|
| 1871 |
+
"min_width": null,
|
| 1872 |
+
"object_fit": null,
|
| 1873 |
+
"object_position": null,
|
| 1874 |
+
"order": null,
|
| 1875 |
+
"overflow": null,
|
| 1876 |
+
"overflow_x": null,
|
| 1877 |
+
"overflow_y": null,
|
| 1878 |
+
"padding": null,
|
| 1879 |
+
"right": null,
|
| 1880 |
+
"top": null,
|
| 1881 |
+
"visibility": null,
|
| 1882 |
+
"width": null
|
| 1883 |
+
}
|
| 1884 |
+
},
|
| 1885 |
+
"c1891461614e45189d47fa3948da6808": {
|
| 1886 |
+
"model_module": "@jupyter-widgets/base",
|
| 1887 |
+
"model_module_version": "1.2.0",
|
| 1888 |
+
"model_name": "LayoutModel",
|
| 1889 |
+
"state": {
|
| 1890 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1891 |
+
"_model_module_version": "1.2.0",
|
| 1892 |
+
"_model_name": "LayoutModel",
|
| 1893 |
+
"_view_count": null,
|
| 1894 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1895 |
+
"_view_module_version": "1.2.0",
|
| 1896 |
+
"_view_name": "LayoutView",
|
| 1897 |
+
"align_content": null,
|
| 1898 |
+
"align_items": null,
|
| 1899 |
+
"align_self": null,
|
| 1900 |
+
"border": null,
|
| 1901 |
+
"bottom": null,
|
| 1902 |
+
"display": null,
|
| 1903 |
+
"flex": null,
|
| 1904 |
+
"flex_flow": null,
|
| 1905 |
+
"grid_area": null,
|
| 1906 |
+
"grid_auto_columns": null,
|
| 1907 |
+
"grid_auto_flow": null,
|
| 1908 |
+
"grid_auto_rows": null,
|
| 1909 |
+
"grid_column": null,
|
| 1910 |
+
"grid_gap": null,
|
| 1911 |
+
"grid_row": null,
|
| 1912 |
+
"grid_template_areas": null,
|
| 1913 |
+
"grid_template_columns": null,
|
| 1914 |
+
"grid_template_rows": null,
|
| 1915 |
+
"height": null,
|
| 1916 |
+
"justify_content": null,
|
| 1917 |
+
"justify_items": null,
|
| 1918 |
+
"left": null,
|
| 1919 |
+
"margin": null,
|
| 1920 |
+
"max_height": null,
|
| 1921 |
+
"max_width": null,
|
| 1922 |
+
"min_height": null,
|
| 1923 |
+
"min_width": null,
|
| 1924 |
+
"object_fit": null,
|
| 1925 |
+
"object_position": null,
|
| 1926 |
+
"order": null,
|
| 1927 |
+
"overflow": null,
|
| 1928 |
+
"overflow_x": null,
|
| 1929 |
+
"overflow_y": null,
|
| 1930 |
+
"padding": null,
|
| 1931 |
+
"right": null,
|
| 1932 |
+
"top": null,
|
| 1933 |
+
"visibility": null,
|
| 1934 |
+
"width": null
|
| 1935 |
+
}
|
| 1936 |
+
},
|
| 1937 |
+
"cc1e631bd9154962bcb42eb6fc234303": {
|
| 1938 |
+
"model_module": "@jupyter-widgets/base",
|
| 1939 |
+
"model_module_version": "1.2.0",
|
| 1940 |
+
"model_name": "LayoutModel",
|
| 1941 |
+
"state": {
|
| 1942 |
+
"_model_module": "@jupyter-widgets/base",
|
| 1943 |
+
"_model_module_version": "1.2.0",
|
| 1944 |
+
"_model_name": "LayoutModel",
|
| 1945 |
+
"_view_count": null,
|
| 1946 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1947 |
+
"_view_module_version": "1.2.0",
|
| 1948 |
+
"_view_name": "LayoutView",
|
| 1949 |
+
"align_content": null,
|
| 1950 |
+
"align_items": null,
|
| 1951 |
+
"align_self": null,
|
| 1952 |
+
"border": null,
|
| 1953 |
+
"bottom": null,
|
| 1954 |
+
"display": null,
|
| 1955 |
+
"flex": null,
|
| 1956 |
+
"flex_flow": null,
|
| 1957 |
+
"grid_area": null,
|
| 1958 |
+
"grid_auto_columns": null,
|
| 1959 |
+
"grid_auto_flow": null,
|
| 1960 |
+
"grid_auto_rows": null,
|
| 1961 |
+
"grid_column": null,
|
| 1962 |
+
"grid_gap": null,
|
| 1963 |
+
"grid_row": null,
|
| 1964 |
+
"grid_template_areas": null,
|
| 1965 |
+
"grid_template_columns": null,
|
| 1966 |
+
"grid_template_rows": null,
|
| 1967 |
+
"height": null,
|
| 1968 |
+
"justify_content": null,
|
| 1969 |
+
"justify_items": null,
|
| 1970 |
+
"left": null,
|
| 1971 |
+
"margin": null,
|
| 1972 |
+
"max_height": null,
|
| 1973 |
+
"max_width": null,
|
| 1974 |
+
"min_height": null,
|
| 1975 |
+
"min_width": null,
|
| 1976 |
+
"object_fit": null,
|
| 1977 |
+
"object_position": null,
|
| 1978 |
+
"order": null,
|
| 1979 |
+
"overflow": null,
|
| 1980 |
+
"overflow_x": null,
|
| 1981 |
+
"overflow_y": null,
|
| 1982 |
+
"padding": null,
|
| 1983 |
+
"right": null,
|
| 1984 |
+
"top": null,
|
| 1985 |
+
"visibility": null,
|
| 1986 |
+
"width": null
|
| 1987 |
+
}
|
| 1988 |
+
},
|
| 1989 |
+
"cd7b84b3c61e49d093df411ccfbc3bb0": {
|
| 1990 |
+
"model_module": "@jupyter-widgets/controls",
|
| 1991 |
+
"model_module_version": "1.5.0",
|
| 1992 |
+
"model_name": "DescriptionStyleModel",
|
| 1993 |
+
"state": {
|
| 1994 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 1995 |
+
"_model_module_version": "1.5.0",
|
| 1996 |
+
"_model_name": "DescriptionStyleModel",
|
| 1997 |
+
"_view_count": null,
|
| 1998 |
+
"_view_module": "@jupyter-widgets/base",
|
| 1999 |
+
"_view_module_version": "1.2.0",
|
| 2000 |
+
"_view_name": "StyleView",
|
| 2001 |
+
"description_width": ""
|
| 2002 |
+
}
|
| 2003 |
+
},
|
| 2004 |
+
"d6cc3547f86c4629a8e8e53592167b43": {
|
| 2005 |
+
"model_module": "@jupyter-widgets/controls",
|
| 2006 |
+
"model_module_version": "1.5.0",
|
| 2007 |
+
"model_name": "HTMLModel",
|
| 2008 |
+
"state": {
|
| 2009 |
+
"_dom_classes": [],
|
| 2010 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 2011 |
+
"_model_module_version": "1.5.0",
|
| 2012 |
+
"_model_name": "HTMLModel",
|
| 2013 |
+
"_view_count": null,
|
| 2014 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 2015 |
+
"_view_module_version": "1.5.0",
|
| 2016 |
+
"_view_name": "HTMLView",
|
| 2017 |
+
"description": "",
|
| 2018 |
+
"description_tooltip": null,
|
| 2019 |
+
"layout": "IPY_MODEL_9db3a6aae167458b8e3c6b785c25f225",
|
| 2020 |
+
"placeholder": "",
|
| 2021 |
+
"style": "IPY_MODEL_cd7b84b3c61e49d093df411ccfbc3bb0",
|
| 2022 |
+
"value": " 408/408 [00:07<00:00, 105.92it/s, Materializing param=vision_model.post_layernorm.weight]"
|
| 2023 |
+
}
|
| 2024 |
+
},
|
| 2025 |
+
"d8ee4430ef8c47bc96d6d7ae2060f9ad": {
|
| 2026 |
+
"model_module": "@jupyter-widgets/base",
|
| 2027 |
+
"model_module_version": "1.2.0",
|
| 2028 |
+
"model_name": "LayoutModel",
|
| 2029 |
+
"state": {
|
| 2030 |
+
"_model_module": "@jupyter-widgets/base",
|
| 2031 |
+
"_model_module_version": "1.2.0",
|
| 2032 |
+
"_model_name": "LayoutModel",
|
| 2033 |
+
"_view_count": null,
|
| 2034 |
+
"_view_module": "@jupyter-widgets/base",
|
| 2035 |
+
"_view_module_version": "1.2.0",
|
| 2036 |
+
"_view_name": "LayoutView",
|
| 2037 |
+
"align_content": null,
|
| 2038 |
+
"align_items": null,
|
| 2039 |
+
"align_self": null,
|
| 2040 |
+
"border": null,
|
| 2041 |
+
"bottom": null,
|
| 2042 |
+
"display": null,
|
| 2043 |
+
"flex": null,
|
| 2044 |
+
"flex_flow": null,
|
| 2045 |
+
"grid_area": null,
|
| 2046 |
+
"grid_auto_columns": null,
|
| 2047 |
+
"grid_auto_flow": null,
|
| 2048 |
+
"grid_auto_rows": null,
|
| 2049 |
+
"grid_column": null,
|
| 2050 |
+
"grid_gap": null,
|
| 2051 |
+
"grid_row": null,
|
| 2052 |
+
"grid_template_areas": null,
|
| 2053 |
+
"grid_template_columns": null,
|
| 2054 |
+
"grid_template_rows": null,
|
| 2055 |
+
"height": null,
|
| 2056 |
+
"justify_content": null,
|
| 2057 |
+
"justify_items": null,
|
| 2058 |
+
"left": null,
|
| 2059 |
+
"margin": null,
|
| 2060 |
+
"max_height": null,
|
| 2061 |
+
"max_width": null,
|
| 2062 |
+
"min_height": null,
|
| 2063 |
+
"min_width": null,
|
| 2064 |
+
"object_fit": null,
|
| 2065 |
+
"object_position": null,
|
| 2066 |
+
"order": null,
|
| 2067 |
+
"overflow": null,
|
| 2068 |
+
"overflow_x": null,
|
| 2069 |
+
"overflow_y": null,
|
| 2070 |
+
"padding": null,
|
| 2071 |
+
"right": null,
|
| 2072 |
+
"top": null,
|
| 2073 |
+
"visibility": null,
|
| 2074 |
+
"width": null
|
| 2075 |
+
}
|
| 2076 |
+
},
|
| 2077 |
+
"db773254517a428684eb679771212f2c": {
|
| 2078 |
+
"model_module": "@jupyter-widgets/base",
|
| 2079 |
+
"model_module_version": "1.2.0",
|
| 2080 |
+
"model_name": "LayoutModel",
|
| 2081 |
+
"state": {
|
| 2082 |
+
"_model_module": "@jupyter-widgets/base",
|
| 2083 |
+
"_model_module_version": "1.2.0",
|
| 2084 |
+
"_model_name": "LayoutModel",
|
| 2085 |
+
"_view_count": null,
|
| 2086 |
+
"_view_module": "@jupyter-widgets/base",
|
| 2087 |
+
"_view_module_version": "1.2.0",
|
| 2088 |
+
"_view_name": "LayoutView",
|
| 2089 |
+
"align_content": null,
|
| 2090 |
+
"align_items": null,
|
| 2091 |
+
"align_self": null,
|
| 2092 |
+
"border": null,
|
| 2093 |
+
"bottom": null,
|
| 2094 |
+
"display": null,
|
| 2095 |
+
"flex": null,
|
| 2096 |
+
"flex_flow": null,
|
| 2097 |
+
"grid_area": null,
|
| 2098 |
+
"grid_auto_columns": null,
|
| 2099 |
+
"grid_auto_flow": null,
|
| 2100 |
+
"grid_auto_rows": null,
|
| 2101 |
+
"grid_column": null,
|
| 2102 |
+
"grid_gap": null,
|
| 2103 |
+
"grid_row": null,
|
| 2104 |
+
"grid_template_areas": null,
|
| 2105 |
+
"grid_template_columns": null,
|
| 2106 |
+
"grid_template_rows": null,
|
| 2107 |
+
"height": null,
|
| 2108 |
+
"justify_content": null,
|
| 2109 |
+
"justify_items": null,
|
| 2110 |
+
"left": null,
|
| 2111 |
+
"margin": null,
|
| 2112 |
+
"max_height": null,
|
| 2113 |
+
"max_width": null,
|
| 2114 |
+
"min_height": null,
|
| 2115 |
+
"min_width": null,
|
| 2116 |
+
"object_fit": null,
|
| 2117 |
+
"object_position": null,
|
| 2118 |
+
"order": null,
|
| 2119 |
+
"overflow": null,
|
| 2120 |
+
"overflow_x": null,
|
| 2121 |
+
"overflow_y": null,
|
| 2122 |
+
"padding": null,
|
| 2123 |
+
"right": null,
|
| 2124 |
+
"top": null,
|
| 2125 |
+
"visibility": null,
|
| 2126 |
+
"width": null
|
| 2127 |
+
}
|
| 2128 |
+
},
|
| 2129 |
+
"dbbd386fb13b4dbcae40301c1bbbbf90": {
|
| 2130 |
+
"model_module": "@jupyter-widgets/controls",
|
| 2131 |
+
"model_module_version": "1.5.0",
|
| 2132 |
+
"model_name": "DescriptionStyleModel",
|
| 2133 |
+
"state": {
|
| 2134 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 2135 |
+
"_model_module_version": "1.5.0",
|
| 2136 |
+
"_model_name": "DescriptionStyleModel",
|
| 2137 |
+
"_view_count": null,
|
| 2138 |
+
"_view_module": "@jupyter-widgets/base",
|
| 2139 |
+
"_view_module_version": "1.2.0",
|
| 2140 |
+
"_view_name": "StyleView",
|
| 2141 |
+
"description_width": ""
|
| 2142 |
+
}
|
| 2143 |
+
},
|
| 2144 |
+
"e74fce8c74c548d285e41413679a080d": {
|
| 2145 |
+
"model_module": "@jupyter-widgets/controls",
|
| 2146 |
+
"model_module_version": "1.5.0",
|
| 2147 |
+
"model_name": "HTMLModel",
|
| 2148 |
+
"state": {
|
| 2149 |
+
"_dom_classes": [],
|
| 2150 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 2151 |
+
"_model_module_version": "1.5.0",
|
| 2152 |
+
"_model_name": "HTMLModel",
|
| 2153 |
+
"_view_count": null,
|
| 2154 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 2155 |
+
"_view_module_version": "1.5.0",
|
| 2156 |
+
"_view_name": "HTMLView",
|
| 2157 |
+
"description": "",
|
| 2158 |
+
"description_tooltip": null,
|
| 2159 |
+
"layout": "IPY_MODEL_6baf2c3128d1425c8f979ed703b0fc6b",
|
| 2160 |
+
"placeholder": "",
|
| 2161 |
+
"style": "IPY_MODEL_2f69578d14bb4ad6967e00931f611f63",
|
| 2162 |
+
"value": "Resolving data files: 100%"
|
| 2163 |
+
}
|
| 2164 |
+
},
|
| 2165 |
+
"edcb5218b9bc4e9d9da375eb93de2829": {
|
| 2166 |
+
"model_module": "@jupyter-widgets/base",
|
| 2167 |
+
"model_module_version": "1.2.0",
|
| 2168 |
+
"model_name": "LayoutModel",
|
| 2169 |
+
"state": {
|
| 2170 |
+
"_model_module": "@jupyter-widgets/base",
|
| 2171 |
+
"_model_module_version": "1.2.0",
|
| 2172 |
+
"_model_name": "LayoutModel",
|
| 2173 |
+
"_view_count": null,
|
| 2174 |
+
"_view_module": "@jupyter-widgets/base",
|
| 2175 |
+
"_view_module_version": "1.2.0",
|
| 2176 |
+
"_view_name": "LayoutView",
|
| 2177 |
+
"align_content": null,
|
| 2178 |
+
"align_items": null,
|
| 2179 |
+
"align_self": null,
|
| 2180 |
+
"border": null,
|
| 2181 |
+
"bottom": null,
|
| 2182 |
+
"display": null,
|
| 2183 |
+
"flex": null,
|
| 2184 |
+
"flex_flow": null,
|
| 2185 |
+
"grid_area": null,
|
| 2186 |
+
"grid_auto_columns": null,
|
| 2187 |
+
"grid_auto_flow": null,
|
| 2188 |
+
"grid_auto_rows": null,
|
| 2189 |
+
"grid_column": null,
|
| 2190 |
+
"grid_gap": null,
|
| 2191 |
+
"grid_row": null,
|
| 2192 |
+
"grid_template_areas": null,
|
| 2193 |
+
"grid_template_columns": null,
|
| 2194 |
+
"grid_template_rows": null,
|
| 2195 |
+
"height": null,
|
| 2196 |
+
"justify_content": null,
|
| 2197 |
+
"justify_items": null,
|
| 2198 |
+
"left": null,
|
| 2199 |
+
"margin": null,
|
| 2200 |
+
"max_height": null,
|
| 2201 |
+
"max_width": null,
|
| 2202 |
+
"min_height": null,
|
| 2203 |
+
"min_width": null,
|
| 2204 |
+
"object_fit": null,
|
| 2205 |
+
"object_position": null,
|
| 2206 |
+
"order": null,
|
| 2207 |
+
"overflow": null,
|
| 2208 |
+
"overflow_x": null,
|
| 2209 |
+
"overflow_y": null,
|
| 2210 |
+
"padding": null,
|
| 2211 |
+
"right": null,
|
| 2212 |
+
"top": null,
|
| 2213 |
+
"visibility": null,
|
| 2214 |
+
"width": "20px"
|
| 2215 |
+
}
|
| 2216 |
+
},
|
| 2217 |
+
"f05122bb2187477ab3f5181938c86a21": {
|
| 2218 |
+
"model_module": "@jupyter-widgets/controls",
|
| 2219 |
+
"model_module_version": "1.5.0",
|
| 2220 |
+
"model_name": "HBoxModel",
|
| 2221 |
+
"state": {
|
| 2222 |
+
"_dom_classes": [],
|
| 2223 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 2224 |
+
"_model_module_version": "1.5.0",
|
| 2225 |
+
"_model_name": "HBoxModel",
|
| 2226 |
+
"_view_count": null,
|
| 2227 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 2228 |
+
"_view_module_version": "1.5.0",
|
| 2229 |
+
"_view_name": "HBoxView",
|
| 2230 |
+
"box_style": "",
|
| 2231 |
+
"children": [
|
| 2232 |
+
"IPY_MODEL_e74fce8c74c548d285e41413679a080d",
|
| 2233 |
+
"IPY_MODEL_3686ac8683c745ee8bc5638d1e962c54",
|
| 2234 |
+
"IPY_MODEL_9049575e797c4bffa86129fd9394f8fd"
|
| 2235 |
+
],
|
| 2236 |
+
"layout": "IPY_MODEL_64363326d62641f3aee10f317d92140e"
|
| 2237 |
+
}
|
| 2238 |
+
},
|
| 2239 |
+
"f09cbd8806bc41808b5594cfb16c1a83": {
|
| 2240 |
+
"model_module": "@jupyter-widgets/controls",
|
| 2241 |
+
"model_module_version": "1.5.0",
|
| 2242 |
+
"model_name": "DescriptionStyleModel",
|
| 2243 |
+
"state": {
|
| 2244 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 2245 |
+
"_model_module_version": "1.5.0",
|
| 2246 |
+
"_model_name": "DescriptionStyleModel",
|
| 2247 |
+
"_view_count": null,
|
| 2248 |
+
"_view_module": "@jupyter-widgets/base",
|
| 2249 |
+
"_view_module_version": "1.2.0",
|
| 2250 |
+
"_view_name": "StyleView",
|
| 2251 |
+
"description_width": ""
|
| 2252 |
+
}
|
| 2253 |
+
},
|
| 2254 |
+
"f4fcf974faad42a08fb6863763eedb81": {
|
| 2255 |
+
"model_module": "@jupyter-widgets/controls",
|
| 2256 |
+
"model_module_version": "1.5.0",
|
| 2257 |
+
"model_name": "FloatProgressModel",
|
| 2258 |
+
"state": {
|
| 2259 |
+
"_dom_classes": [],
|
| 2260 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 2261 |
+
"_model_module_version": "1.5.0",
|
| 2262 |
+
"_model_name": "FloatProgressModel",
|
| 2263 |
+
"_view_count": null,
|
| 2264 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 2265 |
+
"_view_module_version": "1.5.0",
|
| 2266 |
+
"_view_name": "ProgressView",
|
| 2267 |
+
"bar_style": "success",
|
| 2268 |
+
"description": "",
|
| 2269 |
+
"description_tooltip": null,
|
| 2270 |
+
"layout": "IPY_MODEL_18111ed761e146698786fb4a5d0c7051",
|
| 2271 |
+
"max": 408,
|
| 2272 |
+
"min": 0,
|
| 2273 |
+
"orientation": "horizontal",
|
| 2274 |
+
"style": "IPY_MODEL_1e523935a83d434c976f10797cbe2286",
|
| 2275 |
+
"value": 408
|
| 2276 |
+
}
|
| 2277 |
+
},
|
| 2278 |
+
"fba6cbf72ad54e748031d11669b3e31a": {
|
| 2279 |
+
"model_module": "@jupyter-widgets/controls",
|
| 2280 |
+
"model_module_version": "1.5.0",
|
| 2281 |
+
"model_name": "FloatProgressModel",
|
| 2282 |
+
"state": {
|
| 2283 |
+
"_dom_classes": [],
|
| 2284 |
+
"_model_module": "@jupyter-widgets/controls",
|
| 2285 |
+
"_model_module_version": "1.5.0",
|
| 2286 |
+
"_model_name": "FloatProgressModel",
|
| 2287 |
+
"_view_count": null,
|
| 2288 |
+
"_view_module": "@jupyter-widgets/controls",
|
| 2289 |
+
"_view_module_version": "1.5.0",
|
| 2290 |
+
"_view_name": "ProgressView",
|
| 2291 |
+
"bar_style": "success",
|
| 2292 |
+
"description": "",
|
| 2293 |
+
"description_tooltip": null,
|
| 2294 |
+
"layout": "IPY_MODEL_891d641ad7634c9aa5186589f2aba958",
|
| 2295 |
+
"max": 272,
|
| 2296 |
+
"min": 0,
|
| 2297 |
+
"orientation": "horizontal",
|
| 2298 |
+
"style": "IPY_MODEL_aa902a918c79418ea4b4516fb5673fa4",
|
| 2299 |
+
"value": 272
|
| 2300 |
+
}
|
| 2301 |
+
}
|
| 2302 |
+
}
|
| 2303 |
+
}
|
| 2304 |
+
},
|
| 2305 |
+
"nbformat": 4,
|
| 2306 |
+
"nbformat_minor": 4
|
| 2307 |
+
}
|
Chapter 3/naive_packing.png
ADDED
|
Git LFS Details
|
Chapter 3/naive_padding.png
ADDED
|
Git LFS Details
|
Chapter 6/chapter_6_loss_vizualisation.ipynb
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
Chapter 6/cross_attention_batched_loss.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
[1.8203, 1.8281, 1.2578, 1.6641, 1.2188, 1.3281, 1.4609, 1.4375, 1.4922, 1.4609, 1.7266, 1.1875, 1.2109, 1.6719, 1.1641, 1.7656, 1.7109, 1.2344, 1.0, 1.4141, 1.2656, 1.2734, 1.1953, 1.8047, 1.1719, 1.3281, 1.0547, 1.4688, 0.9844, 1.7344, 1.625, 1.6094, 1.2734, 1.1484, 1.7031, 1.3828, 1.5078, 1.6328, 1.5547, 1.7812, 1.0234, 1.4688, 1.2891, 1.375, 1.1484, 1.2578, 1.0078, 1.7188, 1.4297, 1.1797, 1.1484, 1.2344, 1.7344, 1.5391, 1.4766, 1.7969, 1.8281, 1.4141, 1.0859, 1.2422, 1.7891, 1.4844, 1.1406, 1.1172, 1.3281, 1.1328, 1.5547, 1.3438, 1.3438, 1.5391, 1.7578, 1.3047, 1.7734, 1.8672, 1.3594, 0.957, 1.3203, 1.3984, 1.5078, 0.8555, 1.2109, 1.4453, 1.5234, 0.9727, 1.2891, 1.4609, 1.5938, 1.2969, 1.5078, 1.5, 1.5234, 1.3359, 1.3828, 0.9844, 1.4922, 1.2969, 1.3984, 1.2266, 1.5859, 1.6484, 1.0781, 1.3906, 1.3281, 0.8906, 1.4453, 1.5859, 1.6641, 0.8789, 1.3828, 1.5391, 1.375, 1.1797, 1.1797, 1.5781, 1.4453, 1.0625, 0.9766, 1.6562, 1.3828, 1.3047, 1.3516, 1.2422, 1.2109, 1.375, 1.2578, 1.1953, 1.6016, 1.3828, 0.832, 1.1953, 1.2891, 1.8203, 1.1953, 1.2422, 1.9844, 1.4219, 1.3047, 1.1328, 1.4688, 1.3672, 1.6328, 1.3984, 1.3672, 1.5547, 1.6016, 1.6641, 1.1719, 1.2109, 1.0469, 1.625, 0.9883, 1.0, 1.1172, 1.8203, 1.3984, 0.9531, 1.7734, 1.2344, 1.3203, 1.4766, 1.3125, 1.1172, 1.1172, 1.4766, 1.2031, 1.0312, 1.2109, 1.2422, 1.1641, 1.4297, 1.5703, 1.5781, 0.8203, 1.0547, 1.0781, 1.0703, 1.3594, 1.2422, 1.0312, 1.2344, 1.3906, 1.25, 1.6719, 1.1484, 1.4062, 1.1562, 0.9766, 1.1406, 1.2891, 1.4219, 1.0938, 1.3828, 1.3594, 1.1484, 1.2891, 1.3594, 1.3828, 1.0391, 1.0156, 1.2188, 1.3047, 1.1797, 1.4922, 1.3125, 1.4844, 1.2109, 1.5781, 1.5469, 1.3516, 1.2188, 1.3203, 1.4062, 1.6016, 1.3281, 1.4531, 1.5312, 1.0938, 1.0547, 1.3516, 1.1406, 1.0703, 1.5625, 1.2266, 1.5312, 1.25, 1.0859, 1.3672, 0.8672, 1.1484, 1.3672, 1.1484, 1.5156, 1.0781, 1.1875, 1.2344, 1.125, 1.0078, 1.2891, 1.0391, 1.4531, 1.0781, 1.0156, 1.2188, 1.1484, 1.4688, 1.6094, 1.4219, 1.0625, 0.9883, 1.2578, 1.0312, 1.1797, 1.2656, 1.375, 1.4453, 0.8633, 1.3906, 0.7148, 1.0469, 1.3125, 1.0312, 1.1719, 1.1797, 0.9531, 0.9492, 1.25, 1.2422, 1.2344, 1.2656, 1.5156, 1.4766, 1.3594, 1.8125, 1.5547, 0.9062, 1.8828, 1.1172, 1.0469, 1.2109, 0.9609, 1.2344, 1.5, 1.0938, 1.7109, 1.4844, 1.3047, 0.9609, 1.3984, 0.7344, 1.5312, 1.0781, 1.5781, 1.2188, 1.1094, 1.4844, 1.1953, 0.7852, 1.0156, 0.8125, 1.1094, 1.3672, 1.4766, 1.0156, 1.0859, 1.2188, 1.4922, 1.3906, 1.1719, 0.9648, 1.1172, 1.0547, 1.1719, 1.75, 1.1172, 1.2266, 1.3984, 1.4062, 1.2891, 1.2891, 1.5469, 0.957, 1.2969, 1.0, 1.0391, 1.2578, 1.4062, 0.9336, 1.2734, 1.3125, 0.7617, 1.1562, 1.2266, 1.1484, 0.8594, 1.0859, 1.6172, 1.2812, 1.3125, 1.1016, 1.4531, 1.125, 0.8242, 1.6328, 1.0312, 1.3125, 1.4062, 1.3828, 1.6562, 1.0781, 1.4062, 1.6094, 1.5781, 0.8984, 1.0625, 1.3516, 1.3906, 1.2188, 1.1406, 1.3594, 1.4297, 1.3047, 1.0469, 1.2344, 0.9219, 1.0391, 1.5156, 1.2656, 0.9648, 1.0859, 1.4141, 1.2266, 1.1094, 1.5078, 1.2188, 0.8438, 1.5078, 1.1641, 1.3359, 1.2812, 1.2109, 1.3281, 1.1406, 1.1406, 1.0547, 0.793, 1.6328, 1.3359, 1.1641, 1.1484, 0.9531, 1.2266, 1.4609, 1.0938, 0.7773, 1.4531, 1.3281, 1.7422, 1.0781, 1.1641, 0.7383, 1.1641, 1.1797, 1.6562, 1.6328, 1.1953, 1.0938, 1.0156, 1.3672, 1.1875, 1.1562, 0.8086, 0.9141, 1.3281, 1.25, 1.3516, 1.2109, 1.2578, 1.3984, 1.1016, 1.0078, 1.4062, 1.125, 1.1016, 1.2891, 1.1016, 0.8516, 1.6406, 1.7734, 1.2188, 1.2969, 1.4219, 1.2109, 1.1484, 1.25, 1.1641, 1.5156, 1.4922, 0.8438, 1.375, 1.4141, 1.4609, 1.3359, 0.9961, 1.1016, 1.1016, 1.0078, 1.7266, 1.1562, 1.5703, 1.1719, 1.0469, 1.0781, 1.1016, 1.4297, 1.3672, 1.3594, 1.3828, 1.1953, 1.4766, 1.375, 0.7617, 0.9258, 1.4375, 1.4375, 1.1484, 0.9531, 1.3828, 0.9219, 0.9414, 1.125, 1.0078, 1.4219, 1.1641, 1.1953, 1.5312, 1.0859, 1.0859, 0.7773, 1.2266, 1.5703, 1.3516, 0.7617, 0.957, 1.4922, 1.3438, 1.2891, 0.6172, 1.3906, 0.9375, 1.2266, 1.3984, 1.2891, 1.3203, 1.125, 1.4062, 1.125, 0.9961, 1.375, 1.6484, 1.0, 1.0938, 1.0469, 0.9844, 1.3672, 0.7539, 1.1328, 1.2422, 1.4688, 1.0, 1.6953, 1.1641, 0.7656, 1.3906, 0.7383, 1.0703, 1.1016, 1.6094, 1.4141, 1.5859, 1.1641, 1.1953, 0.8945, 1.4688, 1.1484, 0.7656, 1.3047, 1.5156, 0.8438, 0.9336, 1.125, 1.4609, 1.2188, 1.1797, 1.3125, 1.3203, 1.3828, 1.2031, 1.3906, 1.1875, 0.8438, 2.0, 0.9336, 0.8047, 1.1641, 1.1797, 1.2734, 1.4766, 1.0469, 1.5703, 1.0, 1.2266, 1.1641, 1.5, 0.957, 0.6719, 0.8477, 1.2969, 1.0156, 0.9219, 1.2422, 1.6016, 1.4531, 1.0, 1.3672, 1.3906, 1.375, 1.375, 0.957, 1.1797, 0.8672, 1.1641, 0.6562, 1.2812, 1.3359, 1.4688, 1.375, 0.9609, 1.2422, 1.1953, 1.1016, 1.3594, 0.9219, 1.2891, 0.7148, 1.1562, 1.1484, 1.2891, 1.2734, 1.2266, 1.1875, 1.1016, 1.0, 1.0547, 1.6328, 1.6406, 1.5625, 0.9219, 0.9297, 0.9961, 0.8516, 1.4062, 0.918, 1.125, 1.2422, 1.2734, 0.8711, 1.3203, 1.2031, 1.1484, 1.3438, 1.2656, 1.5938, 0.9453, 1.0234, 0.8945, 1.2891, 0.8398, 0.8828, 1.0234, 1.0859, 0.8398, 1.3047, 1.2578, 1.3359, 1.2891, 1.0391, 1.3203, 1.0469, 0.8516, 1.2188, 1.1484, 0.9883, 1.3672, 1.1484, 1.2109, 1.0703, 1.4531, 0.9414, 1.2812, 0.7031, 1.2578, 1.3359, 1.5781, 1.2031, 1.3359, 1.0469, 1.2812, 1.4453, 1.25, 1.2031, 1.1406, 0.9375, 1.1328, 1.3047, 1.0547, 1.2344, 0.832, 1.5859, 1.1875, 1.1406, 1.4219, 1.1328, 1.3906, 1.0859, 1.0625, 0.9141, 1.0, 0.7656, 1.0391, 1.0938, 1.0703, 1.4219, 0.7188, 1.5859, 1.2656, 0.9805, 1.2031, 1.3594, 1.4844, 0.7109, 1.4375, 1.5078, 1.6406, 1.0625, 1.5234, 1.1172, 0.8398, 1.2344, 1.3594, 1.5, 0.9648, 0.9453, 0.8086, 1.1328, 1.3125, 1.3047, 1.1406, 0.9766, 1.3281, 1.2969, 1.0156, 1.0781, 1.3125, 0.8359, 1.4531, 1.1094, 1.3594, 1.1562, 1.1406, 1.2344, 1.2578, 1.1953, 1.2266, 0.8633, 1.3281, 1.375, 1.0781, 1.0781, 1.2422, 1.0859, 1.3828, 0.9492, 1.4297, 1.3438, 1.1953, 0.9766, 0.8633, 1.1953, 0.9766, 0.9453, 1.3594, 0.9531, 1.2266, 1.3125, 1.6719, 1.4844, 1.125, 0.8906, 1.5312, 1.2656, 1.0469, 1.0703, 1.0469, 1.3984, 1.0703, 1.1797, 1.5469, 0.9883, 1.0391, 1.3359, 0.75, 1.4766, 0.8789, 0.8711, 1.0391, 1.4453, 1.0547, 2.0, 1.3828, 1.2656, 1.2969, 1.0469, 0.9609, 0.9727, 1.1094, 1.0703, 1.375, 1.1641, 1.1641, 1.2891, 1.2578, 1.1406, 1.6484, 1.2969, 0.8242, 0.957, 1.2891, 1.1484, 1.5938, 0.7656, 0.8477, 1.3125, 1.2031, 0.9922, 1.4609, 1.2734, 1.3125, 1.0312, 0.7969, 1.1484, 1.2188, 1.1094, 1.1406, 1.2578, 0.8984, 1.0312, 1.4453, 1.1641, 1.3281, 1.2734, 1.5234, 1.0078, 0.7617, 0.8164, 1.3906, 1.2266, 1.3359, 1.1484, 1.3047, 1.2109, 1.1094, 0.7656, 0.8203, 2.0, 1.3906, 1.2891, 1.4531, 1.3438, 1.1016, 1.375, 0.9805, 1.3594, 0.9453, 1.4062, 0.9062, 0.9414, 1.1875, 1.1875, 1.0938, 0.957, 1.2656, 1.3828, 1.1328, 1.4609, 1.2031, 1.2891, 1.7656, 0.625, 1.3438, 1.25, 1.4922, 1.0234, 1.25, 0.9492, 1.4922, 0.9688, 0.9531, 1.2656, 0.7188, 1.3906, 1.3594, 1.3359, 1.2109, 1.2969, 1.0703, 1.5859, 1.0391, 1.6484, 1.3516, 1.0625, 1.1953, 1.4844, 1.4609, 1.2969, 1.1484, 0.875, 1.1406, 0.8047, 1.0391, 1.3125, 1.5469, 1.1094, 1.3281, 1.2266, 1.3125, 0.6641, 1.1016, 1.2266, 1.2109, 0.8633, 1.0, 1.3438, 0.7891, 1.2578, 1.125, 1.0078, 1.0156, 1.25, 1.0469, 1.4062, 1.0391, 1.3281, 1.2812, 1.1953, 1.0938, 1.4844, 1.9297, 1.0547, 1.1094, 1.0625, 1.2344, 1.3125, 0.8672, 1.25, 1.9375, 1.375, 1.1953, 0.9141, 1.3984, 1.3594, 1.5156, 1.1406, 0.9102, 1.2812, 0.7383, 0.918, 1.3438, 1.2266, 0.9844, 1.0859, 1.5078, 0.9062, 0.9609, 1.2188, 1.0781, 1.1406, 1.3125, 1.2109, 1.2031, 1.3359, 1.5469, 1.5156, 0.9844, 1.4375, 0.9727, 1.5938, 1.2109, 1.4219, 1.1562, 1.1484, 1.1953, 1.6094, 1.0703, 1.4766, 1.4453, 0.9219, 1.2422, 1.1406, 1.0781, 1.0625, 0.7617, 1.3203, 1.2734, 1.3125, 1.0625, 1.2344, 0.9961, 0.8516, 0.7031, 1.5312, 1.2344, 0.9883, 1.0625, 1.4375, 0.8086, 1.0078, 1.4531, 0.9844, 1.3047, 0.9531, 1.2812, 1.3984, 1.2656, 0.8867, 0.5, 1.1797, 1.3047, 1.0078, 1.2891, 1.1094, 1.0625, 1.0312, 1.3516, 1.125, 1.1406, 1.7344, 1.4062, 1.0078, 1.125, 1.2344, 0.8359, 1.0234, 0.7695, 1.5, 1.1016, 1.4531, 1.5391, 1.3984, 1.0859, 0.9883, 1.0391, 1.4531, 1.1875, 1.0625, 1.0859, 1.0391, 1.1094, 1.3594, 1.2188, 0.9258, 1.2734, 1.3125, 0.9258, 1.0, 1.0938, 0.9258, 1.0469, 0.9219, 1.1328, 1.3359, 1.1953, 1.4297, 0.9492, 0.9883, 1.3281, 0.7969, 1.125, 0.8555, 1.3203, 1.4141, 1.3125, 0.8906, 1.2344, 1.2188, 1.4844, 1.3125, 1.3047, 1.1875, 1.3438, 1.3828, 1.0859, 1.3125, 1.4453, 1.5391, 1.2734, 0.7617, 0.9844, 0.6133, 0.9414, 1.2656, 1.3047, 0.7461, 1.2031, 0.9023, 1.4609, 1.3047, 0.8086, 1.4922, 1.2031, 1.1016, 1.1953, 0.8828, 1.3281, 1.2109, 1.1172, 1.0078, 1.5312, 1.2031, 0.8594, 0.7734, 0.7344, 1.1328, 0.8906, 1.0625, 1.5234, 1.6875, 1.0703, 1.2656, 0.918, 1.0703, 1.2031, 1.1875, 1.25, 1.0859, 1.2656, 1.3281, 1.1797, 1.1641, 0.6602, 0.8789, 0.9609, 1.1094, 1.0703, 1.3281, 1.4219, 1.4531, 1.3828, 1.0234, 1.0, 0.875, 1.0312, 1.2188, 1.0156, 1.1016, 0.875, 1.1797, 1.1797, 1.4844, 1.1797, 1.1484, 1.125, 1.0781, 1.4688, 0.8477, 1.5312, 1.0078, 1.1641, 1.2109, 0.668, 1.3672, 0.9883, 1.0391, 1.2969, 1.0547, 1.25, 1.2891, 1.1953, 0.8672, 0.7656, 0.9922, 1.3828, 1.7578, 0.8555, 1.0234, 0.8516, 1.0469, 1.7109, 1.25, 0.8359, 1.1953, 1.4141, 1.25, 1.1016, 0.8008, 1.2344, 1.3984, 0.9727, 1.6406, 1.25, 0.9219, 1.2188, 1.4375, 0.9219, 1.2734, 1.1406, 1.6328, 0.8203, 1.0469, 0.8984, 1.1953, 1.3125, 1.2734, 1.0859, 0.9062, 1.1719, 1.0938, 1.1875, 1.4922, 1.1094, 0.9414, 1.2578, 0.9258, 1.1484, 1.2891, 0.9336, 0.9023, 1.3906, 1.0703, 0.9883, 1.3438, 0.9727, 0.793, 0.9102, 1.1406, 0.8711, 1.5625, 1.3906, 1.4609, 1.3125, 1.3828, 1.3984, 1.5625, 1.1016, 1.0703, 1.1875, 1.1406, 1.1484, 0.8633, 1.3203, 1.3438, 1.3984, 0.9102, 1.4062, 1.1172, 1.3516, 1.2266, 1.0, 1.3984, 0.8164, 1.0938, 1.2656, 1.0859, 1.0859, 1.4453, 1.3125, 0.9883, 1.6016, 0.9023, 0.9648, 1.1719, 1.0469, 1.1328, 1.1406, 1.1953, 1.3516, 1.0391, 1.4766, 1.5781, 1.1094, 0.8984, 1.0156, 1.25, 1.3125, 1.2031, 1.3281, 1.2891, 1.0156, 1.2188, 1.2891, 0.8906, 0.9922, 1.0469, 1.0234, 0.875, 1.8828, 1.1875, 0.9648, 1.0547, 1.4141, 1.0703, 0.9062, 1.1641, 1.1406, 1.1094, 1.1953, 0.7852, 1.3438, 1.375, 1.0781, 1.1406, 1.0625, 1.0703, 1.2812, 0.9883, 1.2969, 1.1406, 0.9297, 0.7188, 0.7109, 1.0703, 1.2031, 1.2266, 1.5312, 0.7852, 1.1562, 1.2734, 1.0625, 1.2422, 1.0234, 1.4062, 1.3125, 0.8906, 1.2031, 1.0547, 1.0469, 1.0703, 0.9414, 1.4453, 0.9844, 1.25, 1.0078, 1.3906, 1.25, 1.1797, 1.4766, 0.8867, 1.4141, 1.1094, 1.1406, 1.1797, 0.8672, 1.0625, 1.2812, 0.6719, 1.1172, 1.1641, 0.9922, 1.0078, 1.5078, 1.3281, 1.0391, 0.793, 1.4688, 0.7891, 1.0391, 1.3047, 1.25, 1.1562, 0.8906, 1.1875, 1.0703, 1.1016, 0.9258, 1.0859, 1.3359, 1.1328, 1.1641, 1.1641, 1.2969, 1.2969, 1.1875, 0.9023, 1.3516, 0.832, 1.4688, 0.9023, 1.0234, 0.7969, 0.9961, 0.9961, 0.7031, 1.1562, 0.9648, 1.5781, 1.375, 1.7734, 1.4219, 1.3828, 0.8672, 1.0859, 1.2109, 0.8359, 1.7031, 1.5078, 0.9609, 1.1797, 1.3359, 1.4141, 1.2734, 1.3281, 1.2109, 1.1719, 1.3281, 1.0859, 1.0391, 0.9688, 1.1406, 1.1172, 0.9375, 1.0938, 1.1016, 1.75, 0.9531, 1.0469, 0.9609, 1.2031, 1.0391, 1.1406, 1.0547, 0.8477, 0.8516, 0.9609, 1.2422, 0.9297, 1.375, 1.1172, 1.3438, 1.2422, 0.9102, 1.2344, 1.0312, 1.1484, 1.375, 1.2422, 1.4141, 0.9297, 0.9102, 0.8672, 1.1953, 1.3125, 1.1797, 1.2969, 1.1719, 1.1172, 1.2422, 1.1641, 1.0938, 1.5625, 1.125, 0.793, 0.957, 1.2891, 1.2188, 0.625, 1.4453, 1.2578, 1.0, 1.2656, 0.8984, 0.6172, 1.7422, 1.4375, 1.25, 0.8477, 0.8047, 1.0156, 1.3047, 1.25, 0.9297, 1.4688, 0.9531, 1.4141, 0.9688, 0.8867, 1.2031, 0.8398, 1.5391, 1.2109, 1.1719, 1.2422, 1.0547, 0.9648, 1.5703, 0.875, 1.3984, 0.7227, 1.2656, 1.1797, 0.9297, 1.3047, 1.2188, 1.1016, 1.7656, 1.1719, 1.0938, 1.0703, 1.3594, 0.9609, 1.0391, 0.9023, 1.0547, 1.2734, 1.1797, 1.0938, 1.125, 1.1328, 1.0, 1.1562, 0.9961, 1.0859, 1.0234, 1.2422, 1.0547, 1.1953, 1.1797, 1.2188, 1.4375, 1.2266, 1.0938, 1.1797, 0.6992, 1.2812, 1.3828, 1.0547, 0.9961, 1.5156, 1.0391, 1.3359, 0.9805, 0.9062, 1.1406, 1.1328, 1.3516, 1.0312, 1.0469, 1.3359, 0.9531, 1.1328, 0.8047, 0.9883, 1.2031, 1.0938, 1.2656, 1.1719, 1.2109, 1.7812, 1.3672, 1.1328, 1.1562, 1.0938, 0.9961, 0.9766, 1.0781, 1.1172, 1.5156, 1.1484, 1.5938, 1.1094, 0.9883, 0.8477, 1.4766, 1.3125, 1.2812, 1.0625, 0.8945, 1.7578, 1.0703, 0.9258, 0.9609, 1.25, 1.0625, 0.8984, 1.3438, 0.8945, 1.1406, 1.2578, 1.2734, 0.75, 1.2109, 1.2891, 1.1484, 1.3281, 1.1562, 1.1484, 1.0156, 1.0312, 0.9844, 1.1484, 1.0625, 1.0391, 1.2344, 1.0938, 1.4609, 1.3125, 1.1484, 0.8867, 1.1797, 1.3281, 0.9102, 1.0703, 1.1172, 1.25, 1.0156, 1.2344, 1.2578, 1.2422, 1.4219, 1.2578, 1.0781, 1.0859, 0.8906, 1.0703, 1.2734, 1.375, 1.4688, 1.6172, 1.6406, 0.8906, 1.2734, 1.0156, 1.5469, 1.1719, 1.1484, 0.9727, 1.1484, 0.8789, 1.1172, 0.8438, 1.4609, 0.9023, 1.2188, 1.4688, 1.0703, 1.1953, 1.1875, 1.125, 1.0625, 0.9336, 1.7109, 1.1406, 1.0469, 1.3359, 0.9258, 1.1875, 0.9688, 1.2812, 1.0391, 1.1016, 1.2891, 1.1094, 0.9883, 0.9531, 1.2812, 1.0391, 1.1875, 1.0859, 1.1875, 1.3828, 1.25, 1.7031, 1.1562, 0.9023, 1.2812, 0.8164, 0.8984, 0.7461, 1.0781, 0.8281, 0.7734, 0.7188, 1.3203, 1.2812, 1.0469, 1.2734, 0.8359, 1.0234, 1.2812, 0.9961, 1.0078, 1.5469, 0.75, 1.1016, 1.0547, 1.3438, 1.25, 1.2188, 0.8711, 1.1484, 1.0156, 1.2031, 1.0312, 1.1719, 1.2266, 1.2734, 0.918, 1.2188, 1.3906, 1.0312, 1.4844, 1.1641, 0.9102, 1.3984, 0.6523, 1.0547, 1.2969, 1.2344, 1.2578, 1.1406, 1.125, 1.2578, 0.9766, 1.1719, 0.9805, 1.1953, 0.8945, 1.0703, 1.3203, 1.2109, 1.2734, 1.3594, 1.3594, 0.918, 1.2188, 1.0469, 1.3594, 1.1016, 0.7969, 1.4766, 0.9336, 1.1562, 0.9766, 1.0234, 1.2266, 1.1484, 1.5391, 1.4688, 1.5469, 1.5, 1.1172, 1.3125, 1.3047, 0.8867, 1.1484, 0.9648, 0.9648, 1.1484, 0.8516, 1.1172, 0.9258, 1.1875, 1.2266, 1.0469, 0.8438, 1.1641, 1.125, 1.1562, 1.1953, 1.0234, 0.8047, 0.957, 1.5234, 1.9922, 1.3125, 1.1328, 1.0469, 1.2656, 1.2188, 1.2812, 1.0625, 0.8203, 0.8008, 1.4766, 1.0703, 0.875, 0.8086, 1.3203, 1.0391, 0.9648, 1.0859, 1.0156, 1.3594, 1.3359, 1.0, 1.1953, 1.0469, 1.0469, 1.1797, 0.9961, 1.0938, 1.0391, 1.0703, 0.8633, 0.8516, 0.9883, 1.0625, 1.2266, 1.2891, 1.0234, 1.0859, 0.9297, 1.5, 2.0, 0.9688, 1.0859, 1.3516, 1.4219, 1.2578, 1.2969, 1.1562, 1.1016, 0.9336, 1.6094, 0.9219, 0.9609, 0.7109, 0.9375, 0.793, 0.9805, 1.3828, 1.1562, 1.1875, 1.25, 0.957, 0.7773, 0.6836, 1.6094, 1.1094, 1.1641, 1.2422, 0.918, 1.4297, 1.2109, 1.2422, 1.0234, 1.125, 1.1797, 1.0312, 0.9961, 1.0078, 1.1641, 1.0703, 1.4688, 0.832, 1.2266, 0.8281, 1.4531, 0.8945, 0.707, 0.918, 0.7305, 1.0938, 1.1328, 1.4297, 1.1094, 0.8672, 1.125, 1.2422, 0.9492, 1.3359, 1.0547, 1.4922, 1.2578, 1.0234, 1.1328, 1.1484, 1.2734, 1.5547, 1.4531, 1.4219, 0.7188, 0.5703, 1.1328, 1.3281, 1.0703, 1.5078, 0.8047, 1.1016, 1.3047, 1.3047, 1.125, 1.1484, 1.1797, 1.0469, 0.9805, 1.3984, 0.7812, 1.6797, 0.6211, 1.2188, 0.8984, 0.8203, 1.2109, 0.8945, 1.3516, 0.9922, 1.0703, 0.918, 1.1719, 1.2656, 1.0703, 0.7734, 0.6484, 1.2578, 1.3438, 1.0469, 0.9844, 1.1953, 0.8906, 1.0078, 1.0391, 1.625, 1.0156, 1.3672, 1.1719, 1.0938, 0.6328, 1.1172, 1.0781, 1.0469, 1.1562, 1.0391, 1.2812, 0.668, 1.2266, 1.1172, 1.0391, 0.7422, 0.7617, 0.8125, 0.8711, 1.0781, 1.0938, 1.0, 1.3594, 1.0391, 1.375, 0.8867, 1.0625, 1.7656, 1.2188, 1.0625, 1.1094, 1.1562, 1.1797, 0.9766, 1.2891, 0.7461, 1.1875, 1.3984, 1.0938, 1.3984, 1.0469, 1.3516, 1.2109, 0.7578, 1.0391, 1.0, 1.6406, 0.8672, 1.3203, 1.3281, 1.2031, 1.6797, 1.0781, 1.2969, 0.8359, 0.9219, 1.2422, 1.2344, 2.0, 1.1797, 1.25, 1.2344, 1.1094, 1.2734, 1.0234, 1.3672, 0.9453, 1.1484, 0.8438, 0.8438, 1.1172, 1.1172, 0.8672, 1.2422, 1.3438, 0.8594, 1.0234, 1.1953, 1.4453, 1.3594, 1.1953, 0.8164, 1.25, 1.0312, 0.9844, 1.1484, 0.6328, 1.3438, 1.1953, 1.0703, 1.0859, 1.0547, 1.2734, 1.3906, 1.1406, 1.5, 1.5312, 1.2031, 1.1094, 1.1562, 1.2344, 1.0703, 1.0391, 0.9219, 1.5859, 1.1172, 1.4062, 1.1406, 1.0625, 1.3828, 0.8867, 1.3906, 0.9531, 2.2656, 1.0078, 1.4297, 0.9727, 1.1016, 1.0312, 0.9492, 1.25, 0.9336, 0.8672, 1.0547, 1.2031, 0.8633, 0.8633, 1.0, 1.0, 1.1562, 0.957, 1.2266, 1.1094, 1.1562, 1.1094, 0.8828, 1.1172, 1.0859, 1.3047, 0.8047, 0.7773, 0.832, 0.6914, 1.0547, 1.1328, 1.3906, 0.8867, 1.0469, 1.0391, 0.8047, 1.1797, 1.25, 1.1016, 1.2422, 0.9336, 1.4609, 0.7773, 1.2812, 1.1016, 0.8164, 1.3438, 0.9102, 1.2969, 1.6328, 0.8633, 1.0234, 0.9922, 0.9258, 1.1172, 1.25, 1.1641, 0.9648, 1.0703, 1.4453, 0.7305, 1.2734, 0.9375, 1.2344, 1.25, 1.0859, 1.1016, 1.1953, 1.2969, 1.7578, 1.3906, 1.3594, 0.9297, 1.0938, 1.3125, 0.8398, 0.8594, 1.2266, 1.0469, 1.3203, 1.1641, 1.2422, 0.918, 0.8867, 0.6172, 0.8438, 0.6914, 1.1562, 0.8789, 1.3203, 0.8828, 1.2969, 1.3594, 1.0625, 1.0781, 0.9805, 1.2031, 1.5859, 0.9961, 1.3047, 1.4922, 1.1328, 0.9492, 1.1641, 1.125, 1.1094, 0.9961, 1.1172, 0.7383, 1.0859, 1.6172, 1.1719, 1.3906, 1.1484, 1.2344, 0.8867, 0.6797, 0.9102, 1.2266, 1.5547, 1.0312, 1.125, 1.2109, 1.0078, 0.9258, 1.0859, 1.3984, 1.3281, 0.9883, 0.9688, 1.5078, 0.832, 0.8281, 1.0312, 1.0, 1.1719, 1.1562, 1.0312, 1.5312, 1.2734, 1.0234, 1.0391, 0.7734, 0.8828, 0.7891, 1.4766, 1.3125, 0.9727, 0.9453, 1.1562, 1.0625, 1.4922, 0.9375, 1.2031, 1.5078, 1.2656, 1.2656, 0.9492, 1.1562, 1.4375, 1.1719, 0.8672, 0.8086, 0.6797, 1.2734, 1.2031, 1.1094, 1.2344, 1.375, 0.9336, 1.3359, 1.0156, 0.918, 1.0625, 1.1953, 0.7773, 1.0469, 1.125, 0.582, 1.3359, 0.6914, 1.0078, 1.2656, 1.0078, 1.1406, 1.0156, 0.9453, 1.0703, 0.8945, 0.9062, 0.8945, 1.0, 1.3984, 0.8242, 0.8047, 1.3047, 1.8438, 1.2031, 1.1641, 1.0312, 1.1328, 1.1719, 0.8477, 1.4375, 1.1016, 0.9062, 1.2266, 0.8711, 1.3828, 1.0234, 1.2969, 0.8828, 1.3984, 1.2344, 1.3672, 0.7383, 1.2266, 1.3203, 0.9141, 1.4453, 1.1172, 1.3984, 0.8789, 1.0312, 1.1562, 0.8125, 1.1719, 0.8555, 1.0156, 1.1328, 1.1641, 1.2969, 0.7891, 1.1172, 0.9922, 0.7617, 0.875, 1.3516, 0.8594, 0.5078, 0.707, 1.4219, 0.6992, 1.3906, 1.0469, 0.8906, 0.9922, 1.1328, 1.1172, 1.4453, 0.957, 0.6836, 1.1562, 0.918, 0.918, 1.6875, 1.0469, 1.4609, 0.8438, 1.1016, 1.3203, 0.5586, 1.3047, 1.0312, 1.125, 1.6875, 0.8477, 0.7773, 1.0, 1.0234, 0.9766, 0.9727, 1.125, 1.0859, 1.0156, 0.9102, 1.3594, 1.2188, 1.1953, 1.3516, 1.3516, 1.0703, 1.5859, 1.2578, 1.2422, 1.125, 1.1641, 0.8281, 1.1328, 1.3047, 1.125, 0.9727, 1.0078, 1.2891, 1.1797, 1.2266, 1.1328, 0.9062, 0.9766, 1.2422, 0.957, 1.1719, 1.1875, 1.3984, 1.1953, 1.125, 1.0703, 0.6055, 0.9844, 1.0938, 1.1484, 1.1016, 1.2344, 0.793, 1.2109, 1.3359, 1.0547, 1.1328, 1.5859, 1.0469, 1.0938, 0.8516, 1.1484, 1.2891, 1.2656, 1.0938, 1.3047, 0.9219, 1.0703, 0.8711, 1.4766, 1.2031, 1.1406, 1.0547, 1.2891, 1.3594, 0.9609, 1.3672, 0.7539, 1.1016, 1.2422, 0.9609, 1.1094, 1.1172, 1.1484, 1.4297, 1.7266, 1.3281, 0.8867, 1.0156, 0.9648, 0.8125, 0.7852, 0.9297, 0.9609, 0.9297, 0.9922, 1.1328, 1.1016, 1.1562, 1.0234, 1.1406, 1.2812, 1.4688, 0.7148, 1.3047, 1.1641, 0.9492, 1.0156, 1.25, 1.3047, 1.0703, 0.9766, 0.9492, 1.0312, 1.1641, 1.3047, 0.9961, 1.7422, 0.9883, 0.6641, 1.0078, 1.1562, 1.1094, 1.0938, 0.8086, 1.2969, 1.3281, 1.0156, 1.1797, 1.1719, 0.8008, 1.5625, 1.2422, 1.4375, 0.8633, 0.8672, 0.8164, 1.3828, 0.9102, 0.8555, 1.4531, 1.1406, 0.9844, 1.0938, 0.8828, 1.1172, 0.8164, 1.1328, 1.2344, 1.1094, 1.4844, 1.0859, 0.7891, 1.6406, 1.5156, 1.2031, 1.5547, 1.0859, 0.7773, 0.8438, 0.9414, 0.9023, 1.2891, 1.1797, 1.3906, 1.4531, 1.3359, 1.1797, 1.0234, 0.9727, 1.0234, 1.7656, 1.2344, 1.4219, 0.7188, 1.1719, 1.0078, 0.9023, 1.0078, 1.0391, 1.4531, 1.0, 1.2578, 0.9648, 0.9883, 1.0391, 1.1641, 1.1172, 0.8828, 1.0312, 1.2422, 0.8906, 1.3359, 1.4297, 1.2734, 1.4219, 0.8438, 1.2578, 1.2266, 1.3281, 1.4766, 0.8555, 1.3594, 0.5977, 1.2344, 1.0625, 1.0156, 1.0156, 1.2031, 1.0469, 1.0391, 1.0391, 1.5938, 1.0312, 1.1797, 1.2969, 1.1094, 0.8398, 1.0078, 0.8047, 0.8594, 0.8047, 1.2031, 1.2109, 0.8008, 1.0234, 1.4453, 1.6953, 0.9492, 0.8789, 0.8398, 0.9492, 1.1953, 1.5, 1.0234, 1.3672, 1.3125, 1.0781, 0.7734, 1.3125, 1.1797, 1.1172, 1.0078, 1.5312, 1.1016, 1.0625, 1.0625, 0.9727, 0.9805, 1.1797, 0.9922, 0.8359, 0.9688, 1.4297, 1.3594, 1.0, 0.9688, 1.0312, 1.1406, 1.0234, 1.0781, 1.3203, 0.9883, 1.4922, 1.0312, 1.1875, 1.4062, 1.3672, 0.707, 1.1797, 1.3281, 0.9766, 1.1328, 1.5234, 0.9102, 0.668, 1.0078, 1.1562, 1.0312, 1.2422, 1.0156, 1.1328, 0.9297, 0.7344, 1.2891, 0.5742, 1.6641, 0.7344, 0.8555, 1.2422, 1.0703, 0.9062, 0.832, 1.3125, 1.1797, 0.8242, 1.2578, 1.4531, 1.25, 1.1406, 1.25, 1.5234, 1.2656, 0.9883, 1.0938, 1.1016, 1.0234, 1.2031, 1.25, 1.0156, 1.1328, 1.1797, 1.5469, 0.7812, 1.0, 1.4297, 1.0938, 1.0469, 1.1172, 1.1875, 1.1094, 0.9805, 1.0859, 1.2656, 1.1328, 1.0469, 0.9883, 1.0859, 0.9102, 1.1562, 1.375, 0.793, 0.9375, 1.3906, 1.6094, 0.9805, 1.0, 1.2188, 1.1875, 1.2266, 1.1562, 1.4375, 1.5781, 0.957, 1.2109, 0.9375, 1.6875, 0.8594, 0.9805, 1.2969, 1.0391, 1.0703, 1.125, 1.3984, 1.0312, 1.4141, 1.3281, 1.0781, 0.9844, 1.6406, 1.0625, 1.0312, 1.0703, 0.8594, 1.0078, 0.8711, 1.2109, 1.1016, 1.3281, 1.2109, 1.1641, 1.2031, 1.6484, 0.7969, 0.832, 0.6836, 1.625, 1.5625, 0.7852, 1.1797, 0.9414, 0.7461, 1.3906, 1.2969, 1.3828, 1.2891, 0.9648, 1.1562, 1.0, 1.1172, 0.6133, 1.0234, 0.8633, 1.0547, 0.9102, 1.3672, 1.0703, 1.1641, 1.1016, 0.7969, 0.9805, 1.0859, 1.1094, 1.3906, 1.0547, 1.125, 1.4062, 0.9453, 1.1953, 0.9531, 1.4062, 1.0312, 0.9453, 0.5391, 1.7188, 0.7891, 0.9922, 1.1406, 1.0469, 1.1562, 0.6758, 1.0312, 1.5469, 1.1641, 1.1562, 1.6328, 1.4219, 0.9531, 0.875, 0.9258, 0.6797, 1.2344, 0.75, 0.9805, 1.0, 0.8203, 1.0391, 0.9453, 1.25, 1.0234, 1.2734, 0.7812, 1.0, 1.2422, 1.3828, 1.4453, 1.0547, 1.0156, 1.2969, 1.2812, 0.9414, 1.3594, 1.1641, 1.0469, 1.0781, 1.2031, 0.8711, 0.8672, 1.2812, 1.2031, 1.5781, 1.1328, 1.7266, 0.7891, 1.0859, 1.0703, 1.3203, 1.1797, 1.6641, 1.1172, 0.8281, 1.3594, 1.0, 1.4219, 0.875, 1.1562, 1.2109, 1.3984, 0.6641, 1.0078, 1.2109, 1.2188, 1.2188, 1.1172, 1.0312, 1.0078, 1.0547, 0.9531, 0.7539, 1.0078, 1.2422, 1.0703, 0.9062, 1.0234, 1.0078, 1.2344, 1.1094, 1.0703, 0.9844, 1.2266, 0.8555, 0.7109, 1.0391, 1.1875, 1.1172, 0.9766, 1.25, 1.25, 1.1016, 1.2344, 0.6484, 1.0469, 1.1094, 1.3281, 1.1172, 0.9414, 1.3203, 1.1875, 0.9531, 1.6875, 0.5781, 0.6562, 1.2734, 1.7109, 1.5781, 1.2969, 1.0, 1.0547, 1.0859, 0.8555, 1.3203, 0.7578, 1.0625, 0.9258, 1.0312, 1.3984, 1.0781, 0.9219, 0.9453, 1.0703, 0.9062, 0.8555, 1.3203, 1.0469, 1.3203, 1.0703, 0.7422, 1.0078, 1.2031, 0.8633, 1.0078, 0.7812, 0.9844, 0.957, 1.2188, 0.8906, 1.2031, 0.7812, 0.9453, 1.3047, 1.0938, 1.0703, 0.8125, 1.0469, 0.6172, 0.7422, 0.9766, 1.5078, 1.1016, 1.125, 1.2188, 1.2656, 1.4844, 1.3672, 1.7109, 1.2031, 1.0938, 1.1953, 1.625, 0.8438, 1.0547, 0.9922, 1.1641, 1.1172, 1.2734, 1.1875, 1.3984, 0.7266, 1.25, 0.9922, 0.6562, 1.25, 0.9023, 0.9219, 1.0, 0.9297, 1.0078, 0.8672, 1.2578, 1.0312, 1.5703, 1.3672, 1.5078, 1.0234, 0.8242, 1.0156, 1.125, 1.0938, 0.8516, 1.1797, 1.1172, 1.1641, 1.1484, 0.9062, 1.3672, 0.9297, 1.3984, 1.3047, 1.3594, 1.0234, 1.0, 1.0469, 0.8164, 1.2109, 0.918, 1.0781, 0.6953, 1.5312, 0.7812, 1.1875, 0.707, 1.375, 0.8516, 1.125, 0.9531, 1.7734, 1.1016, 1.0469, 1.1797, 1.0469, 0.8789, 1.3516, 1.6172, 1.0547, 1.1406, 0.832, 0.7656, 0.8477, 0.7578, 1.2578, 0.9688, 1.5625, 1.2031, 0.8945, 1.4844, 1.4062, 1.2969, 1.3281, 1.4531, 1.6172, 0.9609, 1.3906, 0.707, 1.9062, 1.1875, 1.0781, 1.0781, 1.0234, 1.1875, 1.9688, 0.918, 1.3281, 1.5625, 1.0625, 1.1797, 1.3203, 1.0078, 1.0547, 1.0703, 1.1328, 1.0078, 1.3438, 0.8398, 1.0938, 1.0156, 1.25, 1.1719, 0.8945, 1.2656, 1.1875, 0.7891, 1.1875, 1.375, 1.3438, 0.8359, 1.0, 1.3203, 1.3516, 1.1406, 1.0547, 1.3594, 0.8984, 1.0938, 0.8711, 1.2188, 1.1406, 1.3516, 1.0938, 0.957, 0.9609, 0.707, 0.9414, 1.7109, 0.875, 0.7188, 0.9766, 0.9414, 1.2734, 1.0703, 0.9258, 0.9141, 0.9844, 1.0781, 1.2734, 0.8047, 1.0234, 1.3594, 1.3906, 1.2266, 1.3281, 1.2188, 0.9102, 1.1641, 1.0859, 1.1484, 1.2031, 1.1328, 0.9609, 1.0156, 1.25, 1.0469, 1.1953, 1.1016, 1.0391, 0.9297, 0.7305, 1.0234, 1.1797, 1.1484, 1.2031, 1.4062, 0.6953, 1.4453, 1.3672, 0.9102, 1.0703, 1.0156, 1.0469, 1.5, 0.875, 1.2422, 0.9219, 1.2656, 1.1172, 1.0234, 1.2734, 1.2031, 1.3047, 1.0547, 0.8438, 0.7852, 1.0469, 1.0469, 0.8555, 1.3047, 1.4688, 0.625, 1.0938, 1.2812, 1.6641, 0.7891, 0.9531, 1.0, 0.8242, 1.6875, 1.3516, 1.3828, 1.375, 1.1094, 1.5078, 1.1797, 1.25, 1.3672, 1.0547, 1.1953, 1.2812, 1.4844, 0.8789, 1.3047, 1.1406, 0.8086, 0.8086, 1.0625, 0.6133, 0.8359, 0.7539, 0.8594, 1.0781, 1.1172, 1.2422, 1.2344, 1.1719, 1.5625, 0.8477, 1.0391, 1.0781, 1.4688, 1.0547, 1.0938, 1.4297, 0.8633, 0.8906, 1.1094, 0.9805, 0.7617, 1.3125, 1.0859, 1.0391, 1.4453, 0.7266, 0.9922, 1.3828, 0.9961, 1.0547, 1.0625, 1.3047, 0.9531, 0.707, 1.1484, 1.1953, 1.0938, 1.4922, 1.8047, 1.2891, 1.2969, 0.6445, 1.0469, 0.918, 1.0469, 1.3281, 0.8633, 1.0938, 0.8086, 1.0547, 1.3906, 0.8867, 0.8789, 1.0391, 1.3125, 0.9844, 1.0078, 1.3203, 0.7617, 0.7227, 0.918, 1.2812, 1.1484, 1.1875, 1.6562, 1.1641, 0.8438, 1.1719, 1.1016, 1.2188, 1.0234, 0.9102, 0.8164, 1.3672, 0.7383, 0.8555, 0.9336, 1.2344, 1.1797, 1.0156, 0.8789, 0.8008, 0.8438, 0.9805, 0.8633, 1.1484, 1.25, 0.9023, 1.0703, 1.4922, 1.0938, 0.9375, 1.0156, 1.1172, 0.9414, 0.9648, 0.9219, 1.1016, 1.1328, 1.3203, 1.1875, 1.0391, 1.2266, 1.2109, 0.8789, 1.1172, 0.9453, 1.375, 1.2891, 1.0469, 0.9844, 1.2188, 1.1484, 0.9336, 1.1484, 1.2969, 0.9141, 1.0156, 0.9961, 0.9375, 0.9609, 0.9219, 1.2969, 0.7891, 0.707, 1.4922, 1.1328, 0.8164, 0.8633, 1.0078, 1.6016, 1.0859, 0.8594, 1.2422, 0.9844, 1.4219, 1.4922, 1.0234, 1.0312, 0.9883, 1.2109, 1.0, 1.3203, 1.1016, 1.1719, 0.8438, 1.1016, 1.125, 0.6719, 1.0781, 1.25, 1.4609, 1.4219, 0.957, 1.3438, 0.7734, 1.0312, 0.7148, 1.3672, 0.9844, 0.7422, 1.0391, 0.918, 1.4844, 1.2812, 1.0391, 1.1172, 1.0391, 1.0703, 1.5, 1.1016, 0.8828, 1.2734, 1.4922, 0.8203, 1.0625, 1.3438, 0.9805, 0.8672, 1.2031, 1.0781, 1.3906, 1.375, 1.0469, 0.8828, 1.0391, 1.25, 1.0703, 1.2734, 1.3594, 1.0078, 0.9961, 1.3125, 0.9336, 1.1719, 0.8711, 0.8906, 0.7578, 1.2891, 1.1406, 0.9688, 1.2422, 1.1641, 1.1328, 1.4688, 1.0391, 1.3047, 0.6836, 1.25, 0.9688, 1.0312, 1.0703, 1.3516, 1.0703, 1.0078, 1.1641, 1.1016, 1.2891, 1.0625, 1.1562, 0.9609, 1.4297, 0.9922, 1.0781, 0.7578, 1.2344, 1.3281, 1.0938, 0.9141, 1.1016, 1.2656, 1.2266, 0.8438, 1.2188, 0.9297, 1.1406, 0.6484, 1.3047, 1.0391, 1.2031, 1.2656, 0.707, 0.8203, 1.1719, 0.9766, 0.6875, 1.0078, 0.8633, 1.4062, 1.6094, 1.3594, 1.3359, 1.0547, 0.8086, 1.0859, 0.6562, 0.9688, 0.9062, 1.2422, 1.3828, 1.2578, 0.9883, 1.0859, 1.2812, 1.3359, 0.8867, 1.1797, 0.8867, 1.2578, 1.0391, 1.0391, 1.1484, 1.2031, 0.9375, 1.1641, 0.9805, 1.7266, 0.9805, 1.3281, 1.2266, 0.5703, 0.9961, 1.1641, 0.7344, 1.0156, 1.0469, 1.4219, 1.1406, 1.1016, 1.1484, 1.125, 1.2656, 1.4844, 0.8438, 0.9805, 1.3828, 1.3359, 1.3203, 1.1016, 1.2109, 1.2734, 1.6172, 0.9844, 1.0391, 0.8047, 1.4297, 0.8984, 1.0625, 1.1562, 0.7305, 0.7344, 0.9922, 1.3984, 1.0703, 1.0547, 1.2891, 1.0156, 0.9883, 1.0391, 1.3516, 0.8086, 1.1641, 1.0156, 1.1016, 0.7969, 0.832, 0.5586, 1.1406, 1.1406, 0.6953, 0.8008, 1.0547, 1.2812, 0.8945, 1.0703, 1.2812, 1.0781, 0.5078, 1.1094, 1.1875, 1.1016, 1.125, 1.125, 1.0547, 0.6719, 1.1953, 0.7656, 1.125, 0.8359, 1.2266, 1.4219, 0.9219, 1.1016, 0.9961, 1.5391, 0.832, 1.1641, 1.0234, 0.9648, 1.125, 0.918, 0.8516, 0.9258, 1.1094, 1.3906, 0.9141, 1.0703, 1.3203, 0.7656, 1.4609, 1.0703, 1.0, 0.9258, 0.8242, 0.7695, 0.8359, 0.832, 1.2578, 0.8359, 1.0625, 1.4844, 1.1562, 0.8438, 1.1562, 0.9453, 0.8516, 0.7734, 0.7852, 0.9531, 1.0781, 1.6172, 1.2344, 1.125, 1.2578, 0.8711, 1.125, 1.0859, 0.9531, 1.1094, 0.9453, 1.4141, 1.1875, 0.6797, 1.0078, 1.0547, 1.1328, 1.2266, 0.9883, 1.0156, 0.9961, 1.3125, 1.4219, 1.1641, 1.1172, 1.0547, 0.7148, 1.2188, 0.9961, 0.9688, 0.8594, 0.793, 1.4141, 1.0391, 0.9141, 0.5703, 0.9258, 1.3672, 0.8984, 1.0781, 1.0078, 0.9062, 0.793, 1.0781, 0.8945, 1.1406, 1.2969, 0.9844, 1.1641, 1.1094, 1.1406, 1.2031, 1.5391, 1.3359, 1.1094, 0.9219, 1.2578, 0.6562, 1.6484, 1.1406, 1.2422, 1.3359, 0.9219, 1.2031, 1.125, 0.8867, 0.8516, 1.5312, 0.7109, 1.1562, 1.7344, 1.1328, 1.3281, 0.7656, 0.9414, 1.3203, 1.1562, 1.875, 1.0234, 0.8711, 1.3359, 1.125, 1.1484, 1.1016, 1.0625, 0.8398, 1.2656, 1.0625, 1.0469, 1.2344, 0.6875, 0.7969, 0.8047, 1.2344, 1.0312, 1.3125, 1.2344, 1.0547, 0.707, 0.8984, 0.9766, 0.75, 1.1406, 1.1172, 1.1094, 1.4922, 1.2266, 1.2344, 1.1328, 1.2734, 1.3906, 1.375, 0.9023, 0.9922, 1.4141, 1.25, 0.7734, 1.6719, 1.25, 1.2188, 0.6328, 0.8555, 0.9453, 1.0391, 1.3047, 1.2266, 1.0234, 1.1797, 1.1016, 0.707, 1.2812, 1.2656, 1.1094, 0.9648, 1.1016, 0.9023, 0.9883, 1.25, 1.3125, 1.5234, 0.8359, 0.6914, 1.1484, 0.9219, 1.2266, 1.0781, 1.5156, 0.707, 0.9922, 0.9414, 1.1328, 1.1016, 0.8516, 1.0156, 1.6094, 1.0156, 0.8633, 0.7383, 1.3594, 0.7695, 0.7344, 1.1641, 1.3984, 1.5312, 1.0859, 0.875, 1.0156, 0.8398, 0.6797, 1.1641, 1.5078, 0.7812, 0.8945, 1.1406, 1.7578, 1.25, 1.4609, 1.0078, 1.5625, 1.0703, 1.0938, 1.125, 0.9297, 1.3906, 0.8281, 0.8047, 1.375, 1.0234, 1.6875, 1.0703, 1.2344, 0.8789, 1.1172, 1.1562, 1.1328, 1.0781, 1.3906, 0.9844, 1.1562, 1.0234, 1.4453, 1.1016, 1.1328, 1.4219, 1.0234, 0.9609, 1.3203, 1.0938, 1.1953, 1.3047, 0.918, 0.8672, 1.3672, 1.1875, 0.9844, 0.9336, 1.2344, 1.1562, 1.0312, 0.8086, 1.0859, 0.6992, 1.2656, 1.4375, 1.2266, 1.125, 0.9297, 0.8359, 1.1172, 1.3047, 0.8984, 0.9297, 0.9102, 1.2656, 0.543, 0.6758, 1.4688, 1.2969, 1.1719, 0.9609, 1.0938, 0.9219, 0.6875, 1.1094, 1.0781, 0.8945, 0.9258, 1.1016, 1.0625, 1.3828, 0.7969, 1.0234, 0.9453, 1.4453, 1.4844, 0.9375, 0.9922, 0.9883, 1.2969, 1.0469, 1.1328, 0.8672, 1.1719, 1.0078, 1.1875, 0.9844, 1.2266, 0.8594, 0.8867, 1.3828, 1.2344, 1.2188, 1.0625, 1.0469, 1.1719, 0.9805, 1.4453, 1.1094, 1.1406, 0.9844, 1.1875, 1.5078, 1.3281, 1.5078, 1.0312, 1.25, 0.957, 0.8008, 1.2422, 0.7578, 1.3281, 0.9883, 1.1953, 1.3828, 1.3281, 1.1094, 1.0078, 1.4141, 0.793, 1.0625, 1.1484, 0.9414, 0.9766, 1.0234, 1.3203, 1.1875, 0.8125, 1.1484, 0.8906, 1.1406, 0.9531, 0.5938, 1.4062, 1.1094, 1.0859, 1.0391, 1.0547, 0.7383, 1.2578, 0.9844, 1.0391, 1.0859, 1.1406, 0.957, 1.0703, 0.9844, 1.125, 1.0703, 1.3438, 0.7578, 0.7578, 0.6484, 1.1016, 0.9219, 0.9922, 0.7305, 0.8945, 0.7852, 1.0078, 0.9258, 0.9961, 1.2266, 0.6992, 1.1094, 0.9531, 0.8867, 0.9922, 0.7852, 1.1094, 0.8516, 0.875, 1.3828, 1.6719, 1.1797, 0.9766, 1.1172, 1.125, 1.2812, 1.3281, 0.7461, 1.125, 0.9766, 1.1094, 0.8242, 0.9844, 1.0078, 1.0078, 0.9844, 1.0859, 1.1328, 1.0625, 1.3281, 1.2422, 1.0234, 1.0547, 0.9609, 1.4375, 0.9766, 1.3516, 1.5625, 0.8438, 0.8555, 1.2188, 0.9062, 1.1641, 1.1406, 0.9531, 1.0391, 0.7578, 1.0156, 1.0078, 1.0469, 1.1641, 0.9883, 1.0156, 1.125, 1.0703, 1.2812, 1.375, 0.9883, 1.4062, 1.0078, 1.0312, 0.9688, 1.2188, 1.0547, 1.0078, 0.9766, 1.1484, 0.9219, 1.0938, 1.4766, 1.1016, 0.8047, 0.9688, 1.0078, 1.5469, 1.1875, 1.7266, 0.8945, 0.957, 0.8164, 1.1172, 0.6641, 1.6328, 1.2578, 0.8945, 0.7891, 1.5078, 1.2266, 0.8906, 0.918, 1.3281, 1.0, 1.2812, 1.0078, 1.2109, 1.0703, 1.4844, 1.2344, 1.4062, 1.1172, 1.2656, 1.3047, 0.8906, 0.9023, 1.0625, 0.9727, 1.1016, 0.9766, 1.1016, 0.7383, 1.1875, 0.9844, 1.0938, 1.2891, 1.2422, 1.0469, 0.9688, 1.2969, 0.8438, 1.1797, 1.1328, 0.8906, 0.8555, 0.5586, 1.0391, 0.7812, 1.375, 0.7227, 0.8438, 0.9141, 0.9727, 1.0078, 1.1406, 0.9805, 1.0781, 0.7578, 1.0703, 0.8711, 0.6953, 0.9648, 0.9219, 1.3516, 0.8086, 0.9531, 1.0938, 1.0078, 1.1797, 0.8047, 1.5547, 1.6328, 1.1094, 1.1328, 1.4297, 1.2422, 1.0938, 1.5547, 1.0469, 1.4062, 0.8672, 1.2266, 1.0625, 1.1016, 1.1172, 1.1094, 1.3203, 1.0703, 1.1953, 0.9648, 0.7969, 1.2422, 1.0547, 1.0156, 1.0, 1.1562, 1.2812, 1.0781, 1.7812, 0.6875, 1.2656, 1.2969, 1.125, 1.2656, 1.3047, 0.8672, 0.8945, 1.3672, 1.0625, 1.2109, 1.0156, 1.0625, 1.0234, 0.6992, 1.2656, 1.1172, 0.6484, 1.1484, 0.8516, 0.7461, 0.9297, 0.8516, 0.9766, 1.1641, 1.0781, 1.5781, 1.0156, 0.8477, 1.0625, 1.3047, 0.6953, 0.9844, 0.8164, 1.0078, 1.0859, 1.5703, 1.2734, 0.9844, 1.4297, 0.9727, 0.7344, 0.9336, 1.2031, 1.0, 1.2188, 1.0859, 1.3359, 1.2812, 1.3203, 1.0312, 0.7617, 0.8281, 1.3438, 1.0312, 0.8203, 1.1406, 1.2109, 1.3047, 1.1016, 0.9258, 1.1328, 0.9219, 1.1875, 0.9961, 1.5938, 1.3359, 1.3125, 0.9648, 1.6172, 1.0, 0.9258, 1.1641, 0.9336, 1.1172, 0.8477, 1.4297, 1.0781, 0.8359, 1.2578, 0.8359, 1.5, 1.1953, 0.7109, 1.1172, 1.2656, 0.6094, 1.1797, 0.6602, 0.8945, 1.2891, 1.0234, 1.5781, 1.3281, 0.8633, 1.4609, 0.6914, 0.9648, 0.5977, 1.2031, 1.3125, 1.3828, 1.2422, 1.0078, 1.8125, 1.1875, 0.9688, 1.2578, 1.3125, 1.2031, 1.3047, 1.1562, 1.1953, 0.7383, 1.0781, 1.1562, 0.9922, 1.0859, 0.9961, 1.0703, 0.75, 0.625, 1.1953, 1.0938, 1.1484, 0.9648, 0.7148, 1.1406, 1.125, 0.9297, 1.4297, 1.0469, 0.7305, 1.0, 0.8281, 1.3672, 1.3438, 1.3047, 0.8555, 1.7344, 0.9844, 0.9648, 1.0781, 1.3359, 0.7031, 1.0391, 0.7734, 0.9531, 0.8945, 1.1484, 0.6094, 1.5625, 0.9492, 0.9336, 0.8203, 1.0938, 1.2109, 0.7148, 0.7461, 1.0469, 1.5391, 1.0312, 1.0938, 1.1406, 1.375, 0.8398, 1.2031, 1.375, 1.6406, 0.9141, 1.2578, 0.9492, 1.25, 1.0625, 1.1484, 1.0859, 0.9922, 1.1641, 0.6602, 0.668, 0.8711, 1.0469, 0.6875, 0.9219, 1.0703, 0.8633, 1.3906, 0.793, 1.6016, 0.7539, 0.793, 1.1953, 0.9297, 0.7695, 0.8438, 0.6367, 1.375, 1.1484, 1.2266, 1.4453, 1.0703, 0.9609, 1.2734, 1.4297, 1.3516, 0.8125, 1.3594, 1.0703, 1.1328, 1.2734, 0.832, 0.9258, 1.1328, 1.2266, 1.1641, 0.875, 1.0391, 0.8242, 1.2344, 1.2422, 0.8906, 1.1406, 1.2266, 1.0391, 1.0859, 1.1875, 0.9961, 1.1719, 1.1016, 0.7461, 1.2031, 1.4453, 1.0781, 0.8984, 1.3125, 0.8906, 0.9648, 0.8906, 1.2422, 1.2969, 1.1328, 1.6016, 1.1406, 1.1328, 0.9961, 1.1641, 0.6992, 1.5312, 1.3828, 1.1562, 1.0391, 0.9336, 0.7578, 1.6797, 0.8008, 1.0859, 0.8828, 1.125, 1.3438, 0.8555, 1.2109, 0.8281, 0.8086, 0.7969, 1.1172, 1.0156, 0.8438, 0.9492, 1.1484, 0.9492, 1.2422, 0.8398, 1.1953, 0.9727, 0.957, 1.3984, 1.375, 0.9414, 1.2188, 1.3438, 1.0938, 1.1797, 1.1641, 0.8672, 1.0234, 1.0234, 1.1016, 1.0625, 0.7812, 0.8281, 0.9531, 1.6094, 1.3125, 1.1172, 1.0938, 0.8984, 1.375, 1.2578, 0.6055, 0.9961, 1.3281, 0.8711, 0.9453, 1.0078, 1.125, 1.2578, 0.9336, 1.2891, 1.0703, 1.5547, 1.1953, 1.5312, 1.1562, 1.2891, 1.1172, 0.8711, 1.1641, 0.5117, 1.2031, 1.1953, 1.2422, 0.8594, 1.1484, 1.1562, 1.1641, 1.2734, 1.2969, 1.4141, 0.8594, 1.3047, 0.498, 1.4688, 1.2812, 1.0078, 0.9492, 1.0547, 1.2031, 1.1328, 0.7734, 0.8984, 1.1562, 1.4922, 0.9648, 0.957, 1.3281, 1.3125, 0.6992, 0.9297, 0.9727, 0.957, 0.8516, 1.0156, 1.2969, 1.1797, 0.9844, 1.1641, 1.0312, 0.9766, 0.7695, 1.3125, 1.0703, 1.1797, 1.0156, 1.0547, 1.0781, 1.1484, 1.2656, 0.9414, 0.7383, 1.2578, 1.0156, 1.0156, 1.1094, 0.7812, 1.3906, 1.1016, 1.3438, 0.9219, 1.125, 1.2969, 0.8711, 1.6719, 1.5156, 0.9297, 1.0234, 1.2109, 1.5312, 1.0859, 1.0078, 1.0938, 1.125, 0.9062, 1.1328, 1.2656, 1.2734, 1.1797, 0.9336, 0.8398, 1.3359, 1.6875, 0.8672, 1.0234, 1.4062, 1.4766, 1.0781, 0.707, 1.1562, 1.5625, 1.2422, 0.9062, 1.2812, 1.3203, 1.0859, 0.8828, 0.9961, 0.9883, 1.0156, 1.3516, 0.8711, 1.0234, 0.8242, 0.8438, 0.9453, 0.8008, 1.25, 0.9531, 0.7695, 1.1797, 0.9141, 0.7383, 1.0156, 1.2891, 1.2578, 1.2656, 0.9883, 0.957, 1.0469, 0.9102, 1.2734, 1.0156, 1.0625, 1.0938, 1.3281, 1.1484, 1.5859, 1.7344, 1.3047, 0.9102, 1.0781, 1.3047, 1.2969, 1.0859, 1.2344, 1.0312, 1.3125, 1.0938, 1.125, 1.2656, 1.3047, 1.0391, 1.0312, 1.1875, 0.8594, 1.0312, 1.3672, 1.0625, 0.625, 0.9648, 1.3828, 1.4141, 0.957, 1.3047, 0.793, 0.957, 1.5625, 1.0547, 0.7969, 1.1484, 1.0703, 1.0938, 1.1953, 0.8555, 0.9805, 0.8203, 0.7969, 0.8398, 1.0625, 1.1953, 1.2266, 1.0547, 1.3047, 1.1719, 1.2812, 1.0859, 1.2969, 1.1953, 0.9805, 1.1016, 1.0078, 1.2266, 1.0078, 1.5312, 0.9102, 0.8828, 0.9922, 1.0859, 1.2891, 1.0, 1.1172, 0.9219, 1.2188, 1.4922, 1.0703, 1.1719, 0.7969, 1.6016, 1.0469, 1.1016, 0.9297, 0.9414, 1.2031, 1.0703, 1.2344, 1.0078, 1.3125, 0.8906, 0.8086, 1.125, 1.4141, 1.5703, 0.7734, 1.3359, 1.1406, 0.9219, 0.7734, 0.9141, 1.0781, 0.9492, 0.7539, 0.6211, 0.707, 1.25, 0.9375, 1.1953, 1.0859, 1.0, 1.2344, 1.2578, 1.3047, 1.6562, 0.8359, 1.0938, 0.7891, 1.2188, 0.9961, 1.1016, 0.6953, 0.7461, 0.7852, 0.9297, 0.8711, 1.2891, 0.8984, 1.1328, 0.6523, 0.8125, 1.0703, 1.4297, 1.4688, 1.0938, 0.7188, 1.3203, 1.0156, 1.1484, 0.8867, 1.1016, 1.1172, 1.0625, 1.0938, 1.1094, 0.6367, 1.0234, 1.1172, 1.1641, 0.9062, 1.2344, 1.0312, 1.0391, 1.0938, 1.4688, 1.0547, 1.5469, 0.9062, 1.3125, 1.1016, 0.9883, 0.9141, 1.0312, 1.0312, 0.9102, 1.5703, 1.4219, 0.9141, 1.5156, 1.4688, 1.1875, 1.1172, 1.0156, 0.918, 1.0, 1.3906, 1.0469, 1.1094, 1.0859, 0.9531, 0.9766, 0.9453, 0.9336, 1.2109, 0.9141, 0.7656, 0.8398, 1.1562, 1.4297, 1.1094, 1.0625, 1.5938, 0.8203, 1.0391, 0.9727, 1.0938, 1.0234, 1.1406, 0.9141, 1.0547, 0.8047, 1.2344, 0.9688, 1.3203, 1.1797, 1.5859, 0.7969, 1.3672, 1.0, 0.957, 1.3203, 0.9414, 0.8086, 1.3594, 0.8867, 0.8672, 0.9297, 1.1484, 1.3828, 0.875, 1.0938, 1.1562, 1.2578, 0.6875, 1.1406, 1.5938, 0.8828, 0.918, 1.0938, 1.0703, 1.2891, 1.5469, 1.2344, 1.4062, 1.0859, 1.375, 0.8438, 0.9727, 1.4062, 0.9922, 1.5156, 0.8906, 0.8867, 0.8984, 1.5469, 1.1719, 0.8438, 1.2734, 1.2891, 1.1797, 0.9961, 0.8398, 1.4609, 1.1406, 1.375, 1.0078, 1.125, 1.2578, 1.1484, 0.9414, 1.1406, 1.0, 1.1797, 1.0078, 0.9922, 1.2812, 1.3828, 1.2266, 1.0703, 0.9023, 1.3828, 0.7969, 0.8633, 1.2812, 0.8125, 1.0, 1.2656, 1.3828, 1.2344, 1.2969, 1.0391, 1.1328, 0.7891, 1.0781, 1.3047, 0.9453, 1.0703, 1.0156, 0.8672, 1.25, 1.7812, 0.9375, 1.0781, 1.2422, 1.1562, 0.8242, 1.0859, 1.3594, 1.1641, 0.8906, 0.8867, 1.2109, 1.1016, 0.75, 0.7422, 0.7422, 1.1328, 1.0, 1.1875, 0.9805, 0.9062, 1.1797, 1.0547, 1.3438, 0.9883, 1.3125, 0.9688, 1.3359, 1.2109, 1.0469, 1.3047, 1.0703, 1.3125, 0.8047, 0.9922, 1.1719, 1.1562, 0.9141, 0.6758, 1.5234, 0.8945, 1.3516, 1.125, 1.2188, 1.3281, 1.4609, 0.9805, 1.2266, 0.8711, 1.4141, 0.9492, 1.1641, 0.6367, 0.7734, 1.7656, 1.4141, 0.7773, 1.25, 0.9375, 1.5781, 0.9727, 0.918, 1.6328, 1.1484, 0.8594, 0.8398, 1.1094, 1.0625, 1.1406, 0.707, 1.1094, 1.0625, 0.9883, 1.2266, 1.0078, 1.0234, 0.9688, 1.2422, 0.7617, 0.6211, 0.8945, 1.0547, 1.2344, 0.9453, 1.4531, 1.0703, 0.7109, 1.2656, 1.3906, 1.2578, 1.1641, 1.1641, 1.3359, 1.0391, 1.2344, 1.2969, 0.7695, 0.8008, 1.125, 0.875, 1.4297, 1.3984, 0.8984, 1.3047, 1.2422, 0.7891, 1.1719, 0.8633, 1.0312, 1.0547, 1.4297, 0.9453, 1.1484, 1.1328, 1.2344, 1.125, 1.1406, 1.0312, 0.7344, 1.2891, 0.8867, 0.6211, 1.1719, 1.0625, 0.7383, 0.9336, 1.4922, 0.8555, 0.8945, 0.75, 1.0781, 0.793, 1.3828, 1.1328, 1.6172, 1.0312, 1.0938, 1.3047, 1.1094, 0.8398, 0.9531, 0.9844, 1.3672, 1.2422, 1.0469, 0.8672, 1.3594, 1.0078, 1.4766, 1.3906, 1.4531, 1.0859, 1.2578, 1.2734, 1.2344, 1.0469, 0.7969, 1.3047, 0.9766, 1.2266, 0.8477, 1.2188, 1.2344, 1.25, 1.3984, 1.2656, 0.9766, 0.8828, 0.7305, 0.6719, 1.0312, 1.0625, 1.0234, 1.0703, 0.957, 1.0938, 0.7773, 0.9727, 1.3125, 0.793, 1.1719, 1.2969, 1.0859, 1.1328, 1.1172, 1.1484, 1.4766, 0.8164, 1.4062, 1.3828, 0.9922, 0.9336, 1.4141, 1.2812, 1.2578, 1.0156, 0.8164, 0.8477, 1.1562, 1.0625, 1.1797, 1.5703, 1.3125, 0.9453, 1.0781, 1.3438, 0.8594, 1.1953, 1.0391, 1.2734, 1.0391, 1.4844, 1.1484, 1.0469, 1.1797, 1.2656, 1.3203, 1.1562, 1.25, 0.8086, 1.3828, 1.25, 0.8867, 1.0469, 1.6406, 1.2578, 0.8789, 0.8906, 1.3906, 1.2266, 1.0, 1.1016, 1.1016, 1.1719, 1.1875, 0.7656, 0.8516, 0.8086, 0.7422, 0.9141, 1.4062, 1.1875, 1.2578, 1.1641, 0.9297, 0.9258, 1.0547, 0.7656, 1.1641, 1.6094, 1.5703, 0.9336, 0.7734, 0.9727, 1.4297, 1.3125, 0.9062, 1.1797, 0.9883, 1.1172, 1.2188, 1.2109, 1.3672, 1.2578, 1.1094, 0.6836, 0.9062, 0.8203, 0.8164, 1.2656, 1.2812, 0.9062, 0.7695, 1.1406, 1.1016, 0.9336, 1.5469, 1.1484, 1.2422, 0.9102, 1.1562, 1.1562, 1.1094, 1.0781, 1.1562, 1.0312, 0.8984, 1.4688, 1.375, 0.9688, 0.8984, 0.8125, 1.0156, 0.7461, 1.3828, 0.6719, 1.1719, 0.7969, 0.8711, 0.9492, 1.1406, 1.2812, 0.9336, 1.1094, 1.4609, 0.8867, 0.9375, 1.0, 1.2578, 1.2344, 1.0703, 1.0938, 0.9922, 1.5859, 1.2109, 1.1172, 0.6445, 1.1484]
|
Chapter 6/cross_attention_batched_loss_256_image_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
[2.4531, 1.7734, 1.6953, 2.2344, 1.3281, 1.4844, 1.9688, 1.0625, 1.5234, 1.3203, 1.4219, 1.8125, 1.0859, 1.9453, 1.3828, 1.6953, 1.3438, 1.75, 1.3516, 1.3594, 1.1406, 1.7344, 1.5391, 1.4062, 2.3125, 1.6484, 1.2656, 1.4531, 0.8164, 1.6172, 1.3984, 1.2109, 1.1094, 1.4062, 1.3203, 1.1875, 2.0156, 1.5938, 1.2031, 1.3906, 1.8125, 0.7891, 1.4922, 1.4141, 1.4688, 1.2891, 1.7422, 1.4141, 2.3438, 1.4141, 1.3672, 1.5469, 1.0703, 1.3281, 1.8047, 1.6719, 1.2031, 1.4922, 1.7344, 1.6172, 1.7656, 1.7109, 1.2734, 1.9375, 1.1562, 1.1953, 1.5, 1.5781, 1.5391, 1.3359, 0.9766, 1.6484, 1.0703, 1.25, 0.9531, 1.8828, 1.4141, 1.9688, 1.4609, 1.1797, 1.7734, 1.6172, 1.7109, 1.7344, 1.3359, 1.4688, 1.8281, 1.9219, 1.9531, 1.5781, 1.4219, 1.1406, 1.1484, 1.2969, 1.6172, 1.7344, 1.1641, 1.75, 1.3594, 1.0312, 1.2891, 1.3672, 1.125, 1.0625, 1.5703, 1.5391, 1.25, 1.3438, 1.2656, 1.5781, 1.7891, 1.3359, 1.625, 1.7031, 1.25, 1.5, 2.3125, 1.7891, 1.6406, 1.1016, 1.1562, 1.2188, 1.3594, 1.625, 1.6641, 1.3672, 1.0078, 1.4844, 1.0703, 1.625, 1.2266, 1.5, 1.125, 1.1328, 1.1328, 1.3906, 1.5156, 1.7188, 1.4766, 1.3672, 1.1953, 1.6406, 1.3828, 1.2891, 1.8828, 1.4375, 1.4297, 1.3359, 1.1719, 1.4609, 1.0078, 1.0781, 1.75, 1.4219, 0.957, 1.5391, 2.1406, 1.5, 1.6484, 1.8906, 1.1875, 1.1484, 0.9727, 1.7344, 1.1953, 1.2656, 1.0703, 1.7031, 1.6406, 1.5234, 1.4141, 1.9531, 0.9609, 1.1797, 1.3594, 1.4922, 1.4844, 1.125, 1.9531, 1.2188, 1.1641, 1.3125, 1.5, 1.3125, 1.4297, 1.6094, 0.9844, 1.0469, 1.3125, 1.6094, 1.0703, 1.6016, 1.6016, 1.4453, 1.3672, 1.1094, 1.2656, 1.0234, 1.375, 1.6406, 1.1094, 1.2578, 1.3906, 1.1016, 1.8359, 1.5547, 1.3672, 1.2344, 1.0156, 1.3203, 1.3203, 1.5703, 1.6953, 1.4453, 1.1719, 1.4297, 1.2344, 2.4375, 1.2109, 1.6328, 1.4297, 1.4766, 1.2109, 1.0469, 1.5625, 1.2578, 1.3281, 1.7812, 1.3984, 1.1641, 1.2109, 1.375, 1.9141, 1.5312, 1.8281, 1.6562, 1.2344, 1.7344, 1.0312, 1.3594, 1.1406, 0.9883, 1.3906, 1.2109, 1.2266, 1.0078, 1.1719, 1.3594, 1.9844, 1.6719, 1.2266, 1.2891, 1.0703, 1.6406, 1.7812, 1.2109, 0.8828, 1.5781, 1.375, 1.4062, 1.4766, 1.0078, 1.125, 0.9297, 1.2266, 1.5703, 1.4609, 0.9258, 1.1172, 1.3203, 1.0938, 1.3125, 1.375, 0.8867, 1.5625, 1.6797, 1.6328, 1.1172, 1.4375, 1.4609, 0.8477, 0.8047, 1.2031, 1.4375, 1.0547, 1.1719, 1.1172, 1.3359, 1.2969, 1.3281, 1.0391, 1.0703, 0.9844, 1.8516, 1.0469, 1.6016, 1.0312, 1.25, 1.5234, 1.9219, 1.2812, 1.9375, 0.9531, 1.125, 1.3906, 0.9609, 0.7773, 1.5, 1.5547, 1.3438, 1.5938, 1.2031, 1.3359, 1.3906, 1.2031, 1.5938, 0.9336, 1.4844, 1.625, 1.2656, 1.3281, 0.9883, 1.6172, 0.9258, 1.0859, 1.25, 1.3125, 1.3359, 1.2969, 1.2812, 1.5234, 1.2578, 1.3125, 1.5781, 1.5703, 0.9648, 1.4141, 1.5781, 1.8516, 1.4688, 0.9062, 1.4141, 1.2578, 1.3125, 1.2344, 1.8906, 1.9609, 0.875, 1.6328, 1.3203, 1.4844, 1.1406, 1.1953, 1.1719, 1.0391, 0.9102, 1.7109, 0.9609, 1.4688, 0.75, 2.0781, 1.2578, 1.2266, 1.375, 1.8672, 0.7617, 1.2969, 1.1094, 1.1953, 1.3672, 1.3438, 1.0391, 1.0, 0.9531, 1.5938, 1.5156, 1.1875, 1.1094, 1.9141, 1.1016, 1.3828, 0.9414, 1.5, 1.2266, 1.2422, 1.1406, 1.0312, 1.0938, 1.0625, 1.0547, 1.3125, 1.4297, 1.6016, 0.8164, 1.2344, 1.3203, 1.3828, 1.0312, 1.4531, 1.0781, 1.6328, 1.75, 1.6797, 1.4922, 1.3203, 0.7852, 1.1406, 1.4609, 1.0156, 1.125, 0.9102, 1.6562, 1.3359, 1.5391, 0.9414, 1.8281, 1.2266, 0.793, 1.4219, 1.2422, 0.6797, 1.4609, 1.3281, 1.2578, 1.1953, 1.3828, 1.0, 1.0625, 1.1875, 1.2422, 1.0547, 1.1797, 0.5742, 1.2031, 1.1953, 1.375, 0.9922, 1.375, 1.1719, 1.5391, 1.1562, 1.7734, 1.3047, 1.7031, 1.0234, 2.4219, 1.5547, 0.9922, 1.1953, 2.0312, 1.8203, 1.1406, 1.0938, 1.1641, 0.7969, 1.75, 0.7695, 1.2109, 1.1875, 1.3359, 1.3828, 1.2266, 1.0547, 2.0312, 1.1484, 1.6719, 1.1719, 1.6719, 0.9531, 1.1484, 1.2109, 1.0078, 0.7109, 1.375, 1.6953, 0.8867, 1.5625, 1.7031, 1.2344, 1.4141, 1.1406, 0.7969, 1.7812, 1.2578, 1.0625, 1.1172, 0.6836, 0.9766, 1.0547, 0.8242, 0.9102, 0.8047, 2.0312, 1.1719, 0.8828, 2.2969, 0.9609, 0.7227, 1.0, 1.2734, 1.6016, 1.4688, 1.5312, 1.1328, 1.5859, 1.0, 1.4766, 0.9023, 1.1172, 1.4375, 0.8633, 1.4688, 0.8359, 1.6328, 1.6719, 1.2109, 1.0781, 1.4297, 1.3047, 1.2266, 1.5156, 1.2734, 1.2969, 1.5469, 1.0156, 1.6016, 1.1719, 1.6328, 1.1172, 1.3828, 1.1641, 1.0391, 0.6094, 1.3203, 1.2344, 1.3828, 1.4766, 0.9336, 0.7305, 1.2031, 1.3828, 1.2812, 1.3594, 1.0547, 1.5703, 0.6719, 1.4609, 0.793, 1.4062, 0.9062, 0.7695, 1.5781, 1.2578, 1.4844, 1.1641, 1.8828, 0.7539, 1.6953, 1.0859, 1.0547, 1.3203, 1.625, 0.8828, 1.1719, 1.2891, 1.8828, 0.8789, 1.4141, 1.1797, 1.4297, 1.4531, 1.3359, 1.7734, 1.4062, 1.0703, 1.1484, 1.5234, 1.4844, 1.5469, 1.9922, 1.4062, 0.8711, 1.1016, 1.2422, 1.4219, 1.3984, 1.4531, 0.9375, 1.2266, 1.2656, 1.5156, 1.1953, 1.4922, 1.3984, 1.1641, 1.3359, 1.2188, 1.2266, 1.2812, 1.0703, 1.3203, 1.4688, 1.5156, 1.0938, 1.0703, 1.1328, 0.7695, 1.3984, 1.4453, 1.5625, 1.2656, 1.0312, 2.0625, 1.3203, 1.3203, 1.7812, 1.375, 0.8711, 1.4062, 1.3828, 1.1406, 1.5078, 1.2734, 0.9922, 1.4062, 1.1094, 1.2734, 1.0938, 1.2891, 1.2812, 0.9766, 1.8359, 1.4766, 1.2266, 1.2656, 1.5078, 1.25, 1.3281, 0.7617, 0.9883, 1.2031, 1.3828, 1.4844, 1.0938, 1.1406, 1.3594, 0.8594, 1.7656, 0.6953, 1.4297, 1.7344, 1.6953, 0.8008, 1.2344, 1.3047, 0.8438, 1.4141, 1.2422, 1.2578, 1.0156, 2.1094, 1.2578, 1.7266, 1.4141, 1.0547, 1.0312, 1.0859, 1.1406, 1.375, 1.1719, 1.4609, 0.8203, 1.0859, 1.0156, 1.1094, 1.3672, 1.1484, 1.6484, 0.9609, 1.5859, 1.1172, 1.3281, 1.0469, 1.3047, 1.0781, 1.4844, 0.6836, 1.7188, 0.8828, 1.3984, 1.3203, 1.5547, 0.6367, 1.4922, 1.0469, 1.4375, 1.2109, 1.0938, 0.6602, 1.4219, 1.2656, 1.6719, 2.1094, 1.1875, 1.2578, 1.2109, 1.3672, 1.2344, 1.6875, 0.8398, 1.4062, 1.3516, 1.5234, 1.2422, 0.9922, 1.2656, 1.4141, 1.4766, 1.4141, 1.5469, 1.1953, 0.957, 1.0, 1.7656, 1.7344, 1.2188, 1.4453, 1.6328, 1.0938, 0.75, 1.4297, 0.8477, 1.2031, 0.9258, 1.2812, 1.4141, 1.8359, 0.8984, 1.5234, 1.8047, 1.1953, 1.1406, 1.1797, 0.8438, 0.957, 1.2734, 1.1406, 1.4297, 1.4688, 1.5312, 1.125, 1.2031, 1.5234, 1.6484, 1.3984, 0.8242, 1.4219, 1.2266, 1.6484, 1.3516, 0.8633, 0.8945, 1.5469, 1.25, 1.4219, 1.2969, 1.0938, 1.2969, 0.8125, 1.4688, 0.9219, 1.2812, 1.1953, 0.8008, 1.125, 1.2344, 1.9297, 0.7773, 1.6016, 0.9102, 1.2656, 1.0156, 1.5938, 1.0234, 1.4531, 0.8516, 0.9141, 0.7969, 1.2891, 1.1172, 1.8359, 1.3203, 1.2031, 1.5781, 0.6484, 1.1719, 1.5312, 1.5781, 1.2109, 1.3906, 1.2969, 0.8516, 0.8008, 1.5312, 1.1562, 1.1484, 1.1328, 1.1875, 1.1484, 1.5469, 1.5, 1.4609, 1.1484, 0.8945, 1.7422, 1.8047, 0.7812, 0.9531, 1.5547, 1.4922, 1.2422, 1.2344, 1.2578, 0.9492, 1.0156, 1.3672, 1.1484, 1.1875, 0.6992, 1.7188, 0.6328, 1.4219, 0.6836, 1.0469, 1.6406, 1.4531, 1.2578, 1.3672, 1.1094, 1.2109, 0.7422, 1.7734, 0.9453, 1.1797, 1.7031, 1.9531, 1.2656, 1.7109, 1.3828, 1.625, 0.9336, 1.3984, 1.1875, 1.1641, 0.8281, 0.9805, 1.4922, 1.0703, 1.2188, 1.4062, 1.3984, 1.3359, 1.6328, 1.1406, 0.7812, 0.7773, 1.3203, 0.9531, 1.2422, 1.6094, 0.9883, 1.3672, 0.9961, 1.3359, 1.2891, 1.3984, 1.4453, 1.5703, 1.4062, 1.0625, 1.2734, 1.375, 1.3281, 1.1328, 0.8398, 0.832, 1.5859, 1.4922, 0.8281, 0.7539, 0.9336, 1.6719, 1.0781, 1.1094, 1.3594, 1.2344, 1.3438, 1.5469, 1.1016, 1.3594, 1.7656, 0.9609, 1.25, 1.6719, 0.8945, 1.5938, 0.9141, 1.0938, 1.0703, 0.6836, 0.8945, 0.8438, 1.5469, 1.1484, 0.8008, 0.7422, 1.8203, 0.7383, 1.7188, 1.5703, 1.1953, 1.1016, 1.0469, 1.4141, 0.9453, 1.7969, 1.5312, 1.0469, 1.625, 0.9102, 1.0312, 1.2734, 0.918, 0.9844, 0.7969, 1.4219, 1.2656, 1.2344, 1.1875, 1.5234, 1.2734, 1.6172, 0.8086, 1.1641, 1.1562, 1.0859, 1.4375, 1.1484, 1.2031, 0.7422, 1.6641, 0.9727, 1.5312, 1.1875, 0.7539, 0.7227, 1.3516, 0.9766, 1.3672, 1.2344, 1.6406, 1.6016, 1.0078, 1.3125, 0.9414, 1.2891, 1.2188, 1.1484, 1.3047, 0.8047, 1.1953, 1.375, 1.9922, 1.5703, 1.7734, 1.5781, 1.2578, 0.707, 1.0703, 1.1953, 0.9297, 0.8086, 0.9414, 1.5547, 1.2344, 0.9375, 1.0703, 1.0859, 1.2578, 1.4766, 1.6328, 1.1719, 0.9688, 1.7188, 0.9609, 1.3359, 1.2188, 1.2109, 1.4297, 1.3984, 1.1172, 1.625, 1.5312, 1.1953, 0.8359, 1.0781, 1.0078, 0.6992, 1.8359, 0.8594, 0.7461, 1.0078, 0.957, 1.5781, 1.0938, 0.9258, 1.1875, 1.7109, 1.2969, 0.7656, 1.6016, 1.2969, 1.1875, 1.0391, 1.0547, 1.5156, 1.1797, 1.2812, 0.9766, 0.7383, 1.1406, 1.4297, 1.0391, 0.8477, 1.1406, 1.6094, 1.0625, 1.0469, 1.1719, 1.3281, 0.7734, 1.7266, 1.2188, 0.918, 0.8945, 1.2969, 1.2969, 0.9297, 0.7305, 1.0312, 1.4375, 1.7266, 1.3828, 1.5391, 0.9844, 1.7422, 1.1172, 1.3438, 1.0312, 1.3984, 0.8281, 1.4297, 1.5078, 1.4453, 0.918, 1.2656, 1.1719, 1.1641, 1.0547, 0.8672, 1.0, 1.2578, 1.5391, 0.9727, 0.8984, 1.4766, 1.1562, 0.8516, 1.4297, 1.1328, 2.1406, 0.6875, 1.3672, 0.8242, 2.0938, 1.3828, 0.9141, 1.4297, 1.0547, 1.1016, 1.2188, 1.3438, 0.7188, 1.1484, 0.832, 0.8594, 0.9531, 0.875, 0.8242, 1.1406, 0.9648, 1.4453, 1.0234, 1.0391, 1.3672, 1.6953, 1.2734, 0.8281, 1.0938, 1.6719, 1.1719, 1.9453, 0.8516, 0.7656, 1.6797, 1.1562, 1.5547, 1.6172, 1.7109, 1.3125, 0.7891, 1.2109, 1.8984, 0.6758, 1.5156, 0.9922, 1.2422, 1.0469, 1.4062, 1.5078, 1.8594, 1.2812, 0.957, 0.9688, 0.9258, 0.8789, 0.8203, 1.4297, 1.2734, 1.2344, 1.3281, 1.0547, 0.9375, 1.1094, 1.5156, 1.1875, 1.0234, 1.3984, 0.9961, 1.3125, 0.9492, 0.8828, 1.3906, 1.0312, 0.8164, 1.6406, 1.3125, 1.2422, 0.8164, 1.7266, 1.25, 1.1406, 1.1641, 1.8438, 0.9062, 1.3281, 1.1562, 1.0781, 1.2969, 1.4375, 0.875, 0.9531, 1.0312, 1.6641, 1.1875, 1.6016, 0.8633, 1.0547, 1.2734, 1.2578, 1.1172, 1.5625, 1.3828, 1.1875, 0.9102, 1.4844, 1.6953, 1.0156, 1.375, 1.1797, 0.8281, 0.7891, 1.0312, 1.0312, 1.2969, 0.957, 0.9258, 1.1562, 1.3672, 1.1172, 0.8203, 1.3203, 1.0469, 1.1094, 1.9219, 1.5547, 1.5781, 0.9844, 0.8008, 1.1172, 1.5547, 1.2656, 0.9492, 1.8516, 0.957, 1.0234, 1.0547, 1.1016, 1.125, 1.3438, 1.3594, 1.1016, 0.9922, 1.1328, 1.7656, 1.0859, 0.8008, 0.9336, 0.9414, 1.8359, 0.7109, 0.6641, 1.1562, 1.6016, 0.8477, 0.7656, 1.0, 1.2031, 0.918, 1.1797, 1.3828, 1.1875, 2.2188, 1.4375, 1.4062, 1.3672, 1.625, 1.3828, 0.8789, 1.1562, 1.0938, 1.1172, 0.8008, 1.1016, 1.1953, 1.0078, 1.2578, 1.5, 1.0156, 1.125, 1.0547, 1.2812, 1.875, 1.0703, 1.1562, 1.2578, 1.5078, 1.4531, 1.4375, 1.3281, 0.9492, 0.8672, 1.1797, 1.3438, 1.0625, 1.2031, 1.5156, 1.5156, 0.8164, 0.8359, 0.8359, 1.2891, 1.3281, 1.1172, 1.3281, 0.793, 1.2734, 1.8281, 1.5078, 1.6719, 1.0625, 1.0391, 0.9141, 0.7773, 1.1562, 0.8438, 1.6797, 0.9883, 0.7734, 1.5859, 1.2188, 1.1797, 0.9141, 1.2891, 1.4844, 1.3594, 1.1719, 1.1875, 1.4609, 0.9258, 1.4453, 1.5625, 1.1328, 0.8164, 0.9453, 0.6328, 0.9141, 1.8828, 0.9766, 1.5547, 1.5312, 1.0156, 1.0391, 1.3281, 0.8711, 1.1562, 1.2891, 1.1172, 0.9844, 0.6719, 0.7891, 1.7969, 1.7031, 1.5312, 1.3828, 1.1562, 1.7578, 1.2656, 1.3984, 1.3672, 1.4844, 1.0312, 1.2969, 1.6016, 0.8633, 1.5469, 1.5469, 1.5156, 0.8828, 1.1016, 0.957, 1.1562, 1.4688, 0.7422, 1.4141, 0.9492, 0.9805, 1.0234, 1.3828, 1.2109, 1.3125, 1.3594, 1.1406, 1.4141, 1.3516, 1.4609, 1.5312, 1.3672, 0.707, 1.2891, 1.2969, 1.2031, 1.4609, 1.1797, 1.7266, 1.0547, 0.9258, 1.4766, 1.5938, 0.8438, 1.4375, 0.9883, 0.8125, 1.5156, 1.1016, 0.6406, 1.5312, 1.0, 1.3594, 1.4062, 1.3203, 1.4062, 1.7031, 1.3203, 0.9609, 1.4922, 1.1094, 1.0859, 1.6641, 1.6094, 1.3672, 1.4297, 0.9492, 1.1875, 1.125, 1.7422, 1.2969, 1.8438, 1.1406, 0.9922, 1.2188, 1.25, 1.0469, 1.3203, 1.0938, 0.6719, 0.957, 1.1719, 1.1172, 1.75, 1.3906, 0.9453, 1.3281, 1.3672, 1.4297, 1.1562, 1.2656, 1.0703, 0.5586, 1.1719, 1.2344, 1.0312, 1.5078, 1.125, 1.0547, 0.8555, 1.3203, 0.8867, 1.5859, 0.7852, 1.0469, 1.4297, 1.125, 1.3828, 0.9727, 0.9297, 0.7734, 1.3203, 1.2891, 1.2812, 1.3359, 1.3594, 1.1953, 1.5625, 1.125, 1.7188, 1.1641, 1.2266, 1.2422, 0.9883, 1.5, 1.375, 2.2656, 1.3359, 0.7422, 1.2734, 1.1641, 1.1562, 0.8125, 1.4531, 1.3047, 1.3125, 0.9102, 0.8008, 1.0547, 1.6406, 1.3438, 2.2188, 1.3984, 1.3516, 0.8359, 1.0703, 1.1406, 1.0859, 1.5547, 1.5938, 1.1875, 1.0625, 0.9844, 1.3828, 1.0156, 0.6562, 0.6406, 1.4297, 1.3281, 1.4219, 1.25, 1.3359, 0.8203, 0.9219, 1.4141, 1.4609, 1.4062, 0.8867, 1.2031, 0.9609, 1.2812, 1.0312, 1.2734, 1.1094, 1.0078, 1.7266, 0.8945, 1.25, 1.3594, 1.2891, 1.3281, 1.6328, 1.0781, 1.4922, 1.4297, 1.2188, 1.125, 1.1875, 1.1641, 1.4453, 1.4062, 1.4688, 1.2812, 1.5391, 1.0078, 1.5078, 0.7578, 1.6641, 1.1328, 1.2734, 1.3516, 1.1797, 1.2578, 1.3984, 1.7188, 1.3359, 0.9727, 0.9531, 1.2422, 1.3047, 0.9062, 1.3359, 0.9453, 0.8398, 1.3047, 0.6875, 0.9727, 1.4141, 1.4609, 0.9023, 1.5, 1.5859, 0.8945, 1.1719, 1.3828, 0.8828, 0.957, 0.7734, 1.2656, 1.3281, 1.2188, 1.4062, 0.8359, 1.1953, 0.8867, 1.5078, 1.3672, 0.707, 1.1094, 0.9531, 1.6875, 1.1406, 1.2188, 1.4297, 0.8438, 1.2422, 1.2422, 1.3125, 1.4219, 1.6094, 1.1562, 0.8906, 0.8828, 0.7773, 1.2109, 0.9688, 1.6406, 1.0547, 0.918, 1.7734, 0.9492, 0.9219, 1.125, 1.4453, 0.6406, 1.3984, 1.4141, 1.5, 1.0156, 1.1562, 0.7305, 1.625, 1.3828, 2.0469, 0.9688, 1.2969, 0.6758, 1.1172, 1.1875, 1.3125, 1.1953, 0.6523, 0.9102, 1.1484, 0.7188, 0.9336, 1.3594, 1.0703, 1.125, 1.9766, 1.0078, 1.5625, 1.625, 1.2812, 1.1719, 1.0234, 0.6992, 1.1094, 0.9805, 1.1484, 1.7188, 1.1719, 1.4844, 0.9414, 1.2188, 1.0781, 1.0938, 1.125, 1.2578, 1.1406, 1.0781, 1.6172, 0.8125, 1.3516, 0.8789, 1.3906, 1.0703, 1.2266, 0.957, 0.8867, 1.2578, 1.1094, 0.6562, 1.1406, 0.7383, 1.4375, 0.8672, 1.0547, 1.1641, 1.5859, 0.9648, 0.8984, 1.5703, 1.3359, 1.1562, 1.1641, 1.0078, 1.1641, 0.7383, 1.5547, 0.7305, 1.0078, 1.375, 1.5703, 1.3359, 1.4219, 1.4062, 0.6445, 1.4453, 1.3203, 0.7852, 1.3984, 1.25, 1.9141, 1.6094, 1.0703, 1.6484, 1.4219, 0.9297, 1.3047, 1.4375, 1.5156, 1.3203, 0.8398, 1.2812, 1.1328, 1.4219, 1.6328, 1.4766, 1.5703, 1.3516, 0.8945, 0.6367, 1.2266, 0.8086, 0.8438, 0.8867, 1.7188, 0.8672, 1.2891, 1.1406, 1.1172, 1.0312, 1.3438, 0.8281, 1.125, 1.2031, 1.4609, 1.375, 0.707, 1.2344, 1.6016, 1.1406, 1.1641, 1.2578, 1.0234, 1.1016, 1.4844, 0.7422, 0.8945, 1.5078, 1.1094, 0.8281, 1.5703, 1.3281, 1.2734, 0.9023, 1.3281, 1.7656, 1.0625, 0.918, 0.8906, 0.8984, 0.8828, 1.2266, 0.9219, 0.9805, 1.0078, 1.6719, 1.5938, 1.2969, 2.125, 0.8242, 1.2891, 1.3281, 1.1953, 0.9297, 0.9375, 1.1094, 1.8281, 0.6094, 1.4922, 1.3828, 1.1172, 1.2109, 1.1484, 1.3047, 1.2422, 1.4297, 1.25, 1.2266, 1.3516, 0.7852, 0.957, 0.8203, 1.2188, 1.0859, 1.1719, 1.0859, 1.3516, 1.4688, 1.1406, 1.7578, 1.2891, 1.3594, 1.3984, 1.4844, 0.7773, 1.1172, 1.3359, 0.7266, 0.9062, 1.2656, 0.9336, 1.2656, 0.625, 0.9219, 1.5781, 0.6992, 1.2031, 1.1484, 1.2812, 0.9961, 1.8984, 1.0078, 1.5938, 1.0078, 1.0859, 1.1562, 0.8711, 1.0234, 1.2109, 1.3672, 1.1641, 1.0391, 1.5312, 1.4453, 0.9219, 1.1953, 1.2031, 1.2656, 1.2578, 0.6836, 0.707, 1.2734, 1.375, 0.793, 1.3984, 0.9375, 1.1641, 1.2891, 1.2266, 0.9648, 1.3359, 1.0781, 1.0, 1.4219, 1.3281, 1.0234, 1.0078, 0.6797, 0.7461, 1.25, 0.8047, 1.3125, 1.6484, 1.0391, 1.1094, 0.9453, 1.2969, 0.8242, 0.9531, 0.7461, 1.1719, 1.3516, 1.6406, 1.1016, 1.3594, 0.7617, 1.2031, 1.1562, 1.0938, 1.5703, 1.1641, 1.1094, 1.3203, 1.0859, 1.0156, 1.3047, 1.4531, 1.1797, 1.0156, 1.2734, 1.4766, 1.5234, 1.4844, 0.8477, 1.1719, 1.1562, 1.4453, 1.3672, 1.0391, 1.1641, 1.0078, 1.4766, 2.4062, 0.9648, 1.0781, 1.0625, 0.832, 1.1641, 1.1484, 1.2969, 1.4297, 1.0391, 1.0625, 1.4141, 1.3828, 0.9453, 1.2031, 1.3984, 1.2031, 0.9336, 1.1797, 1.0625, 1.7656, 1.0, 0.8203, 1.1172, 1.4297, 1.1172, 1.4844, 1.3828, 1.1406, 1.3594, 1.0938, 0.6641, 1.0938, 1.2969, 1.2578, 0.9102, 1.2109, 1.1562, 0.9375, 1.4141, 1.2266, 1.2578, 0.7422, 1.1562, 1.3672, 0.8281, 0.918, 1.0469, 1.6562, 1.5859, 1.4219, 1.4141, 1.3672, 1.1328, 1.5469, 1.2969, 1.5703, 1.2969, 1.6953, 1.2344, 1.6094, 1.0547, 0.8594, 1.0078, 1.375, 0.9609, 1.0, 1.4297, 0.8711, 1.2188, 0.7305, 1.25, 1.6719, 1.1641, 1.0938, 1.6641, 1.0859, 0.8164, 1.4375, 1.2812, 1.2734, 1.0391, 1.0859, 1.0469, 1.5234, 1.0078, 1.1484, 1.2422, 0.8359, 1.4453, 0.7188, 1.1406, 1.2891, 0.8984, 1.7109, 1.1406, 1.5781, 1.2734, 1.2969, 1.6172, 0.8359, 1.2422, 1.5391, 1.1328, 1.25, 0.9258, 1.1328, 1.1328, 1.2812, 1.6797, 0.8594, 1.0312, 1.1016, 1.2109, 1.1875, 1.4609, 1.0781, 0.8438, 1.2656, 1.6562, 2.0938, 1.3438, 0.8555, 1.1406, 1.0312, 0.8555, 1.1406, 1.4219, 1.7656, 0.6055, 1.3594, 1.375, 1.1875, 1.2031, 1.4609, 1.3281, 0.6055, 1.2188, 1.3672, 1.2031, 1.3359, 1.1406, 0.8477, 1.1797, 1.0781, 0.957, 1.1953, 1.2109, 2.0625, 1.2344, 1.3984, 0.9844, 0.9844, 0.8672, 1.3047, 1.3281, 1.4531, 0.8906, 0.7734, 1.2266, 1.2266, 1.1016, 1.1797, 1.1641, 1.0703, 1.0859, 1.0312, 1.0625, 0.9062, 1.375, 1.2109, 1.3359, 1.4062, 0.9922, 1.0156, 1.1953, 1.2188, 0.8438, 1.4219, 0.7344, 1.7812, 0.9414, 1.0469, 0.9062, 1.4922, 0.9141, 1.2266, 1.1953, 0.7617, 0.9453, 0.4902, 0.8008, 1.0391, 1.1719, 1.3828, 1.2734, 1.2422, 1.5, 0.9688, 0.8555, 0.9297, 1.3672, 1.5156, 1.4062, 1.0234, 1.2344, 1.2891, 1.4922, 1.2266, 1.5781, 1.0078, 1.0625, 1.3203, 1.0859, 1.0938, 0.9688, 1.1641, 1.1719, 0.8398, 0.9023, 1.2109, 1.5078, 0.9844, 1.3828, 1.2188, 0.7812, 1.0234, 1.4062, 1.3906, 1.3594, 1.1094, 1.5703, 1.6016, 1.0703, 0.8945, 0.8438, 1.7578, 1.0547, 1.375, 0.9883, 1.4766, 0.7305, 1.1094, 0.8828, 1.0, 1.2344, 1.4219, 0.5977, 0.8789, 1.2812, 0.9102, 1.2969, 1.0234, 0.9141, 1.0703, 1.4141, 1.4688, 0.9727, 1.2969, 0.9961, 0.793, 1.2188, 1.4219, 1.1328, 0.6211, 1.1484, 0.9219, 1.3438, 1.2031, 0.9375, 1.5625, 1.7188, 1.1016, 0.6484, 1.4609, 0.8672, 1.3047, 1.0312, 1.2969, 0.875, 1.0391, 0.9062, 1.1562, 1.0703, 1.5547, 1.0234, 0.8359, 1.2266, 1.4297, 1.0, 0.9375, 1.5078, 1.3984, 1.5469, 1.0234, 1.2266, 1.0703, 1.4922, 1.1328, 1.1172, 1.1328, 1.0469, 0.7422, 1.2031, 0.7148, 0.8555, 0.9375, 1.2266, 0.793, 1.0391, 0.6211, 0.9453, 0.8828, 1.7031, 0.6172, 1.3516, 1.4297, 1.1172, 1.7266, 1.7734, 1.4219, 1.6016, 1.25, 1.3125, 1.0078, 1.2656, 0.7422, 1.3984, 1.2109, 1.6797, 1.7188, 1.2109, 1.8125, 0.8125, 1.0312, 1.2734, 1.2656, 1.5938, 1.2344, 1.3906, 1.125, 1.5156, 1.2578, 1.3438, 1.1094, 1.1328, 0.9727, 1.4219, 1.5234, 0.9922, 1.0703, 0.9414, 1.3906, 1.0, 1.0703, 1.2031, 0.8398, 1.2734, 0.9102, 1.2109, 1.3984, 1.4766, 1.9453, 1.1953, 0.9219, 0.6172, 1.6016, 0.9961, 1.0938, 1.25, 1.1406, 0.6797, 1.3438, 0.9922, 0.6914, 0.9141, 1.2344, 0.6875, 1.5859, 1.0, 0.6992, 1.4766, 1.4219, 0.9102, 0.9844, 1.5, 1.375, 1.1797, 1.1719, 0.8125, 0.9727, 0.9453, 1.625, 1.3359, 0.9531, 1.5234, 1.2031, 1.2812, 1.1719, 1.4062, 1.3438, 1.0703, 0.7031, 0.9375, 1.2031, 0.6953, 1.4688, 1.3203, 1.1562, 1.4453, 1.125, 1.5391, 1.5391, 1.3672, 1.0938, 0.9805, 1.3984, 0.9961, 1.3281, 1.4766, 2.0312, 1.1797, 0.9258, 1.3906, 0.918, 0.6172, 1.3281, 1.1406, 0.9219, 1.5469, 0.8516, 0.832, 0.9727, 1.7656, 1.1641, 0.8203, 1.2188, 1.6094, 0.8086, 0.7852, 1.7188, 1.5703, 1.2109, 1.4297, 1.0547, 1.4766, 0.7852, 0.9727, 0.8438, 1.0391, 0.8359, 1.0547, 1.5469, 1.2734, 1.6016, 0.707, 1.0625, 1.5625, 1.3906, 0.8398, 1.3047, 1.1094, 1.5, 0.9102, 0.8516, 1.0469, 0.832, 0.9727, 1.4141, 1.2344, 1.7969, 1.2031, 1.125, 1.3047, 1.2344, 1.2422, 0.8945, 0.9102, 1.5234, 1.7266, 0.9883, 0.8438, 1.2266, 1.6406, 0.9336, 0.9883, 1.2109, 1.0, 1.3047, 1.2031, 1.0078, 1.6797, 0.9297, 1.5625, 0.9023, 2.1094, 1.5625, 1.0859, 1.1875, 0.9375, 1.3359, 0.9258, 1.2969, 1.2344, 1.0078, 1.1875, 1.1172, 0.7617, 1.125, 1.5625, 1.0859, 1.0391, 1.3516, 1.0469, 1.3125, 1.1484, 1.1953, 0.6836, 1.0625, 1.2031, 1.3125, 0.7969, 1.6875, 1.2344, 0.6562, 1.375, 1.4219, 0.8867, 1.4062, 1.1094, 0.6836, 1.9219, 1.1484, 0.918, 1.6484, 1.25, 1.5469, 0.875, 1.0703, 0.9219, 0.7578, 1.4062, 1.0703, 0.9375, 1.8594, 0.8047, 1.1641, 1.1797, 1.5156, 1.1562, 1.1094, 1.0938, 1.3281, 0.9258, 1.0859, 0.9141, 1.0938, 1.2656, 1.6875, 1.1406, 1.0234, 1.0469, 0.9297, 1.6328, 1.0625, 1.2734, 0.7578, 1.1328, 1.1328, 0.8125, 1.2109, 1.3828, 0.9844, 0.6133, 1.5781, 1.25, 0.8828, 1.6953, 1.2422, 0.9258, 1.7656, 1.7812, 1.3203, 1.75, 1.3438, 0.8281, 1.1094, 0.9922, 1.2109, 1.0312, 1.3984, 0.8984, 1.2422, 1.0781, 1.4922, 1.2188, 0.8867, 1.6172, 1.7578, 0.8633, 1.25, 1.1094, 0.7188, 1.0312, 1.1484, 1.6406, 1.2266, 1.3984, 1.4531, 1.0938, 0.8672, 1.4375, 1.6797, 1.0625, 0.9219, 1.0156, 1.1016, 0.9453, 0.8633, 1.4062, 1.2109, 1.0078, 0.832, 1.25, 1.5312, 0.9062, 1.3047, 1.0547, 0.8672, 1.7344, 1.1953, 0.8984, 0.8164, 0.9883, 1.4609, 1.4688, 0.8711, 1.2188, 1.125, 1.3438, 1.0, 1.3438, 1.0078, 1.2812, 1.375, 1.0391, 1.2812, 0.6641, 1.0156, 1.0391, 1.2422, 1.0391, 1.1094, 1.0, 1.2891, 1.1172, 1.4531, 0.6602, 1.7578, 1.3438, 1.375, 1.4141, 1.0078, 1.2578, 0.6992, 1.4922, 1.0078, 1.0156, 1.5547, 1.1641, 1.2031, 1.3359, 1.0547, 1.4688, 1.0234, 1.0547, 1.5703, 1.2344, 1.6562, 1.2734, 1.3281, 1.0781, 1.3828, 1.3984, 0.9375, 1.4141, 0.8398, 1.1484, 1.1719, 0.7148, 1.5703, 1.2188, 1.5703, 1.3594, 1.6562, 1.5859, 1.3984, 1.7891, 1.1719, 0.7852, 1.3125, 1.2578, 1.3516, 0.9258, 1.7109, 1.1797, 0.9609, 1.1172, 1.0859, 1.1406, 1.3203, 1.2109, 0.707, 1.0938, 1.0234, 0.9375, 0.9258, 0.9766, 1.2656, 1.7031, 0.9102, 0.9141, 1.0938, 1.6641, 1.4062, 0.9844, 1.0703, 1.3906, 0.8516, 1.25, 1.4766, 0.8867, 1.1016, 1.0, 1.0625, 1.5547, 1.3984, 1.1719, 1.3281, 1.0547, 1.5625, 0.6875, 1.1641, 1.1406, 1.2344, 1.1484, 1.5781, 0.9297, 1.25, 1.1719, 1.2188, 1.3203, 0.8867, 1.2109, 1.2344, 0.8203, 0.9883, 1.0078, 1.8203, 1.2266, 0.8125, 1.125, 0.9336, 1.0234, 1.0938, 1.3672, 1.3359, 1.4688, 1.5938, 1.4609, 1.7656, 1.0078, 0.7266, 1.4297, 1.1562, 1.0703, 0.9492, 0.7539, 0.9844, 0.9922, 0.6719, 0.7734, 1.0703, 0.8086, 0.7695, 0.9219, 1.5156, 1.2812, 1.4062, 0.8125, 1.5156, 0.6758, 0.6055, 0.9688, 1.5312, 1.1172, 0.8867, 1.1641, 1.1953, 1.2578, 0.8242, 0.9336, 1.625, 1.2031, 1.0703, 1.5391, 1.4453, 1.1719, 1.0156, 1.7031, 0.875, 0.8438, 1.2812, 0.9219, 1.1562, 1.3594, 0.7969, 1.1641, 0.9727, 1.3828, 1.1172, 1.6875, 1.1328, 1.1172, 1.2969, 1.2188, 1.1172, 1.1484, 1.5156, 1.1562, 1.1328, 1.2969, 1.4531, 1.3516, 0.7266, 0.6602, 1.3125, 1.3516, 1.1875, 1.2266, 1.0703, 1.2656, 1.4062, 1.0078, 0.8828, 1.4609, 0.8438, 1.25, 1.1719, 1.125, 1.0859, 1.3203, 0.8438, 0.7852, 0.9805, 1.125, 1.3047, 1.3906, 1.0312, 0.9219, 1.2812, 1.5234, 1.3516, 1.3828, 1.4922, 0.8242, 1.4688, 1.1016, 1.1953, 1.0547, 1.3594, 1.0938, 0.9805, 0.7656, 1.4922, 1.5312, 0.9844, 1.2031, 0.832, 1.1328, 1.1562, 0.9375, 1.1641, 1.4688, 1.0, 1.125, 1.4922, 1.6719, 1.4375, 1.4844, 1.0703, 1.9766, 1.3906, 1.4844, 1.3359, 1.2891, 1.3047, 1.1484, 0.8086, 1.1562, 1.0156, 1.2812, 1.0469, 1.25, 1.2578, 0.9531, 0.7266, 0.7969, 1.4062, 1.1016, 0.7344, 1.4844, 0.7812, 1.1953, 1.2422, 1.125, 0.9883, 0.8125, 1.1328, 0.9219, 1.2188, 1.2188, 1.0469, 1.2422, 0.9648, 1.2891, 1.0156, 0.8906, 1.5078, 1.4141, 1.8438, 1.4609, 1.3672, 1.3906, 1.1094, 1.0, 1.0469, 1.2812, 1.4062, 0.8008, 1.6328, 1.0469, 1.1484, 0.6367, 0.7109, 0.9414, 1.5312, 1.1484, 0.9453, 1.0312, 1.1328, 0.918, 1.0469, 1.5234, 1.1016, 1.3438, 0.8203, 0.9336, 0.8633, 1.3281, 1.6875, 1.0469, 1.2266, 1.375, 0.9375, 1.1094, 1.0938, 1.1719, 0.9492, 1.25, 1.1172, 1.0859, 1.4297, 0.8984, 0.6133, 1.3516, 1.25, 0.9766, 0.8047, 1.0703, 0.9727, 0.7891, 1.1016, 1.0312, 0.9961, 1.5547, 0.9648, 1.3984, 0.9531, 1.0703, 1.1562, 1.0234, 1.2266, 1.6094, 2.2812, 1.2109, 0.8438, 1.1328, 1.4844, 1.5859, 1.0078, 1.6875, 1.0078, 1.6016, 1.0703, 0.9297, 1.3203, 1.0312, 1.0625, 1.2891, 0.9023, 1.4844, 1.5703, 0.8945, 1.2812, 0.6875, 0.9258, 0.668, 1.1562, 0.707, 0.8906, 0.625, 1.6484, 1.2422, 1.4609, 1.0547, 1.0391, 1.4844, 1.1641, 0.957, 0.9805, 0.8945, 0.8086, 0.8828, 1.0078, 1.6719, 1.6094, 1.2188, 1.125, 1.6328, 1.4141, 0.7617, 0.9414, 1.7812, 1.0703, 1.0703, 1.5156, 1.0312, 1.0156, 1.0, 1.1484, 0.9688, 1.1406, 1.1797, 0.9688, 1.0625, 1.0859, 1.0781, 1.4844, 1.1562, 1.5234, 1.0, 0.8945, 1.1172, 1.1562, 1.0, 1.3359, 1.2188, 0.75, 0.8438, 0.8633, 0.7812, 0.8828, 0.8906, 0.8281, 1.0703, 1.4453, 0.8164, 1.4922, 1.4219, 1.0781, 1.1094, 1.0078, 0.6914, 1.1641, 1.3516, 1.1484, 0.7891, 1.0625, 1.5156, 0.9297, 1.3906, 1.3438, 1.5859, 1.1328, 1.0, 0.9492, 1.125, 1.0859, 1.2188, 1.3203, 1.1797, 1.4219, 1.5938, 1.7344, 1.2656, 0.8867, 0.7422, 1.0938, 1.5156, 1.3359, 0.8594, 0.9688, 1.8594, 1.0156, 0.9922, 1.1719, 1.1484, 1.6094, 1.2812, 0.9688, 1.3516, 0.6992, 1.3438, 0.9844, 1.2578, 0.9258, 1.1406, 1.4141, 0.9844, 1.7109, 1.5469, 0.6797, 0.875, 1.125, 1.2891, 0.9297, 0.8125, 1.1562, 1.3047, 0.9766, 1.125, 1.3047, 1.0703, 0.9648, 0.7109, 1.0859, 1.1719, 1.3438, 1.0469, 1.4375, 1.2734, 0.5391, 0.6914, 1.1094, 1.1562, 1.375, 1.0703, 1.3125, 0.7578, 1.0625, 1.2656, 0.8555, 0.7969, 1.1094, 1.2109, 0.9844, 1.6641, 1.4531, 0.8711, 1.3438, 1.3594, 0.875, 1.6172, 1.1953, 0.9023, 1.4531, 1.2031, 0.8672, 1.3438, 1.2969, 0.8086, 1.5234, 0.9141, 1.4062, 1.2891, 1.0938, 0.7227, 1.0391, 1.3672, 1.25, 1.125, 0.8086, 0.8203, 1.1562, 0.6914, 0.9375, 0.918, 1.2266, 0.9531, 1.1016, 1.1953, 1.2656, 1.5703, 0.8945, 1.8047, 0.9688, 1.0078, 1.2266, 0.9297, 1.4453, 1.9141, 1.1875, 0.9141, 1.4219, 0.9414, 1.3047, 1.1875, 1.2109, 0.707, 1.2734, 1.5312, 0.7773, 0.8555, 1.2031, 1.125, 1.4297, 1.4453, 1.0703, 0.9961, 1.2266, 1.5547, 1.4453, 1.6641, 1.0859, 0.9062, 1.5625, 0.8164, 1.2578, 0.8125, 1.2422, 1.6562, 1.4688, 0.75, 1.125, 1.0469, 1.3359, 1.1719, 1.2266, 2.0781, 1.0547, 1.6797, 1.25, 1.1406, 0.7305, 1.0781, 0.582, 1.1797, 1.3203, 1.1406, 1.4453, 1.6484, 0.7695, 1.5859, 1.2422, 1.3125, 1.2344, 1.2109, 0.8672, 1.5156, 1.2031, 1.1562, 1.0469, 0.7461, 1.8672, 1.5156, 0.8203, 1.2578, 0.9961, 1.0234, 0.7227, 1.0625, 1.2422, 1.3828, 0.8203, 0.6719, 1.7656, 1.4453, 0.9805, 0.9688, 1.1172, 1.4453, 1.125, 1.8984, 1.7578, 0.9531, 1.3828, 1.0625, 0.8789, 1.2266, 1.2891, 0.8164, 1.1016, 0.9727, 0.9297, 1.4766, 0.7266, 0.9297, 1.8828, 1.3516, 1.0, 0.7852, 0.9453, 1.3906, 0.7539, 1.375, 1.5078, 1.25, 1.0781, 1.5078, 1.4297, 1.9297, 1.1328, 1.3438, 1.1328, 0.9492, 1.1094, 1.4219, 0.9883, 1.2422, 0.8984, 1.2734, 1.0391, 0.8438, 1.0156, 1.3594, 1.7891, 1.2266, 1.5078, 1.4141, 0.8984, 1.75, 0.832, 1.2422, 1.4844, 0.9648, 1.0, 1.2109, 1.2344, 1.3125, 0.7539, 2.5, 0.6914, 1.6406, 1.0547, 1.6328, 0.9922, 0.7734, 1.0625, 0.9727, 1.2969, 0.8906, 1.2578, 0.9805, 1.1328, 0.7734, 1.3828, 1.2891, 1.1328, 1.0234, 0.5469, 1.2578, 1.2109, 0.9297, 1.0234, 1.3906, 0.9258, 0.8945, 1.2031, 1.4609, 1.1719, 1.0156, 1.4375, 0.832, 1.1641, 0.9492, 1.0312, 1.0312, 1.2109, 0.8867, 1.3828, 1.5234, 1.0859, 0.6094, 0.7812, 0.9453, 0.5312, 1.0703, 0.8242, 0.9023, 1.2266, 1.3203, 1.3359, 1.3359, 1.2344, 0.7266, 1.0234, 1.5234, 1.1562, 0.6992, 1.1953, 1.1016, 1.3516, 1.0859, 1.1328, 1.1562, 1.25, 1.0547, 0.8125, 1.5781, 1.2422, 0.8086, 1.0234, 1.4922, 1.1719, 0.918, 0.9688, 1.0391, 1.0391, 0.9805, 1.2734, 1.0547, 2.0156, 0.7812, 0.9453, 1.1406, 1.0078, 0.9961, 0.7422, 1.125, 0.7422, 1.1094, 1.2891, 1.2656, 1.5781, 0.8984, 0.9883, 1.0078, 1.2578, 0.8203, 1.7422, 0.9727, 0.9766, 1.1406, 0.9609, 1.1406, 1.1328, 1.4766, 1.0078, 1.4141, 1.0391, 1.3047, 1.0391, 1.0703, 1.2422, 1.3984, 1.9531, 1.3984, 1.5078, 1.2734, 1.1016, 1.1094, 1.3438, 0.9414, 1.3047, 0.9375, 0.6484, 1.1016, 0.9141, 1.5781, 1.3281, 1.0, 1.1562, 1.4922, 1.1328, 1.7266, 1.1094, 0.8867, 1.0469, 0.668, 0.8828, 1.0469, 0.5742, 0.9414, 1.2812, 0.8086, 1.6484, 0.9531, 0.8125, 1.2734, 1.2031, 1.4062, 1.3672, 1.1953, 1.0625, 1.1641, 1.0859, 1.8203, 1.5625, 1.1484, 1.0078, 1.3672, 1.2344, 1.0938, 1.6953, 1.1562, 1.0781, 1.1641, 1.0703, 0.9609, 1.1641, 1.4219, 0.9102, 1.2656, 1.2656, 1.1328, 0.875, 0.9727, 0.875, 1.0547, 1.2734, 2.0156, 1.4062, 1.1406, 1.4453, 1.1094, 1.1484, 1.0938, 1.3125, 1.2031, 0.8867, 0.707, 0.9102, 0.9414, 1.1797, 1.5312, 1.3594, 1.2031, 1.3438, 0.8281, 1.3359, 1.0078, 1.1562, 1.2734, 1.0391, 1.0, 0.668, 1.4453, 1.3594, 1.5781, 0.9844, 1.4375, 0.9336, 0.7383, 1.2891, 1.6406, 1.3984, 0.7734, 1.0547, 0.6953, 0.7188, 1.2734, 1.0234, 1.0703, 1.25, 0.9531, 1.7031, 1.0859, 1.0156, 1.7344, 1.2812, 1.375, 1.3516, 0.668, 1.4219, 0.7695, 0.9688, 0.7539, 0.7031, 1.1719, 1.7969, 1.1719, 1.2734, 1.2031, 0.7188, 0.9453, 0.8555, 1.625, 1.1016, 1.0234, 1.9688, 0.9844, 1.0781, 1.1641, 1.0938, 1.7188, 1.0938, 1.5781, 1.4141, 0.8516, 1.2266, 0.9961, 1.1562, 1.8906, 0.9961, 1.1328, 1.0391, 0.5781, 1.1328, 1.0078, 0.5625, 1.0078, 0.8086, 1.6797, 0.9102, 1.1953, 1.1641, 1.5, 1.3594, 0.9609, 1.4531, 1.1406, 1.2734, 0.793, 1.4141, 0.8398, 0.75, 1.7344, 0.8672, 0.9453, 1.1797, 1.1328, 0.8125, 0.6797, 1.7344, 0.7578, 0.8828, 0.9961, 1.2344, 1.1875, 1.2188, 0.9883, 0.918, 1.2344, 1.2812, 0.8203, 1.2188, 1.0078, 1.1484, 0.7109, 0.7852, 1.0078, 1.1016, 1.1094, 0.8594, 1.5469, 0.8086, 1.1094, 0.9375, 1.4219, 1.4297, 1.8438, 1.1406, 1.3906, 1.0703, 1.1562, 1.0547, 1.2266, 1.3047, 1.1094, 1.2422, 1.2969, 1.6875, 1.4375, 0.7734, 1.5859, 0.7266, 1.0938, 1.4219, 1.0703, 0.7305, 1.3594, 1.3281, 1.2188, 1.0547, 1.2656, 1.5391, 1.1094, 1.5391, 1.2422, 1.0625, 0.9648, 0.8672, 1.3203, 1.375, 1.0312, 1.0703, 1.1641, 1.3438, 1.0625, 1.4766, 1.2656, 1.0625, 0.8828, 1.2734, 0.8516, 1.1875, 0.8672, 1.1562, 1.0703, 0.9961, 1.3047, 0.918, 0.7734, 1.3828, 1.3125, 1.0391, 1.3281, 1.125, 0.8047, 0.957, 0.918, 0.9844, 0.7266, 0.8281, 1.2656, 1.375, 0.7656, 0.8047, 0.7148, 0.9531, 1.2344, 0.832, 0.8555, 1.3516, 1.0781, 1.1484, 1.0078, 0.9453, 0.9883, 1.0391, 1.3438, 1.2109, 1.2031, 1.7422, 1.0234, 1.0312, 0.6836, 1.125, 0.9414, 1.2578, 0.8359, 1.5391, 1.8203, 1.5078, 0.9297, 1.4219, 1.5078, 0.957, 0.9258, 1.625, 0.8867, 0.8672, 1.7109, 1.1797, 0.7773, 1.6016, 0.7109, 1.7812, 1.375, 0.8867, 0.9336, 0.9258, 0.8867, 1.2734, 0.7422, 1.0078, 0.9258, 1.1016, 1.2031, 0.9414, 1.5703, 1.0625, 0.7227, 1.1719, 0.9531, 1.3906, 1.3359, 1.4453, 0.9766, 1.25, 1.125, 1.2812, 1.7344, 0.9023, 1.0547, 1.4453, 1.5938, 1.1562, 1.2812, 1.0234, 1.5625, 0.7422, 1.7891, 0.9727, 0.8594, 0.8984, 1.1953, 1.0547, 1.2891, 1.3438, 0.9336, 1.4844, 0.8242, 0.8516, 1.2578, 1.1328, 1.4141, 0.9258, 1.6719, 0.832, 0.9062, 0.9805, 0.9688, 1.6172, 1.2734, 0.9844, 1.3203, 1.2891, 1.2109, 1.0781, 1.5781, 1.6172, 0.793, 1.1719, 1.2109, 0.9219, 0.6484, 1.1406, 1.4297, 0.8867, 0.8945, 1.3516, 1.3203, 0.957, 1.3125, 1.0703, 0.6641, 1.2266, 1.1562, 1.1797, 1.3359, 1.0781, 1.3516, 1.25, 1.4531, 0.9727, 0.9258, 0.875, 1.25, 1.0469, 1.3047, 0.9883, 1.6641, 0.9219, 1.3203, 1.3516, 0.832, 0.832, 1.2422, 1.3516, 1.25, 1.3203, 1.0469, 1.3594, 1.1719, 1.3594, 1.2812, 1.2891, 1.7734, 0.8398, 0.9297, 1.2812, 1.6406, 0.7695, 1.0391, 1.0625, 1.2422, 1.1094, 1.0391, 1.1484, 1.125, 0.8398, 1.3828, 1.2109, 0.9844, 1.5391, 1.4375, 1.7656, 1.4141, 1.375, 1.1719, 0.9688, 0.7344, 0.7969, 1.2812, 0.7891, 1.0469, 1.0391, 1.0234, 0.7188, 0.7031, 1.0547, 0.8789, 1.2969, 1.0547, 1.0703, 1.4531, 1.2031, 1.0469, 1.1953, 1.6797, 0.6953, 1.4844, 1.5625, 0.6758, 1.4766, 1.3516, 1.2266, 1.0703, 1.1719, 0.75, 1.2109, 1.2891, 1.2344, 1.3516, 1.0078, 1.2812, 1.1875, 0.75, 1.0, 1.1484, 0.7578, 1.0, 1.1641, 1.1094, 1.3281, 1.0391, 1.3516, 1.9375, 0.9492, 0.918, 0.6562, 0.9375, 1.0078, 1.0859, 1.0156, 0.9414, 1.3281, 1.3906, 0.9766, 0.7812, 0.8281, 1.9688, 1.1719, 0.918, 1.1875, 1.1406, 1.2656, 1.1406, 0.707, 1.7031, 1.5781, 1.7266, 1.0391, 1.6719, 0.875, 0.918, 0.9062, 1.2031, 1.5703, 0.9531, 0.8516, 0.8281, 1.1328, 1.4219, 1.2266, 1.2578, 0.7812, 1.2969, 0.9336, 1.1484, 0.7773, 1.5, 0.7969, 0.8477, 0.9609, 1.2266, 0.957, 1.3359, 1.3203, 1.25, 1.1016, 1.4219, 1.2812, 0.8477, 0.7969, 1.7109, 1.7266, 1.4141, 1.1953, 1.5156, 1.5859, 1.4062, 0.957, 0.7148, 1.125, 0.5508, 1.2812, 0.7891, 1.1562, 1.1719, 1.2969, 1.5547, 0.9258, 1.4531, 1.4219, 1.2734, 1.5625, 1.1094, 1.3047, 1.2891, 0.9219, 0.6602, 1.3359, 1.0156, 1.3438, 1.3984, 1.3359, 0.957, 1.2969, 0.793, 1.2812, 1.3594, 0.9375, 0.9961, 0.9297, 1.3906, 0.9727, 1.8203, 0.8203, 1.1953, 1.4844, 1.1484, 1.3047, 1.1797, 0.7305, 1.3125, 1.1797, 1.3984, 0.9922, 0.7852, 1.4219, 0.793, 0.6328, 1.6797, 0.9375, 0.7266, 2.0625, 1.2266, 1.5781, 1.1797, 1.4844, 1.1797, 1.0859, 1.0, 1.2188, 1.2109, 1.375, 1.3359, 1.4453, 1.0156, 1.6641, 0.9492, 1.5469, 0.7344, 0.9414, 1.125, 1.0781, 0.957, 1.125, 1.4688, 1.4453, 1.0625, 1.0469, 0.9844, 1.2109, 1.1172, 1.4219, 1.75, 0.8281, 0.9727, 1.4609, 1.1094, 1.4688, 0.9141, 1.125, 0.957, 0.9258, 0.9453, 0.7031, 0.9336, 0.6875, 1.125, 0.793, 0.8516, 1.5234, 1.375, 1.1562, 0.75, 1.1875, 0.9219, 0.9648, 1.0, 1.2031, 0.6602, 1.1562, 0.9023, 0.8008, 1.0781, 1.125, 1.4844, 0.8711, 1.2578, 1.3516, 1.1172, 1.4922, 1.7656, 1.5, 0.8477, 1.0781, 1.4766, 0.5898, 1.6641, 0.8477, 1.2812, 1.1953, 1.3906, 0.7148, 1.2422, 1.3984, 1.4531, 0.9414, 1.1016, 1.3984, 0.9375, 0.957, 0.8516, 1.0547, 0.9375, 1.8594, 0.7422, 1.2656, 0.9062, 1.2266, 0.9219, 1.4219, 1.7266, 1.25, 0.7773, 1.1953, 1.0391, 0.793, 1.7734, 0.793, 1.1016, 0.9648, 0.8633, 1.3672, 0.9219, 1.2891, 0.957, 1.6875, 1.6484, 0.7109, 1.3906, 1.3047, 1.0703, 1.8359, 1.0, 1.3281, 1.2109, 0.5312, 0.8125, 1.3516, 1.3516, 1.4375, 1.2266, 1.1484, 1.1016, 0.9805, 2.1406, 0.9297, 0.7773, 0.8945, 0.8672, 1.6406, 0.8945, 1.0625, 1.3906, 1.3906, 0.9531, 1.4297, 0.6562, 1.1016, 0.9219, 0.7617, 1.6172, 0.8438, 1.2422, 1.3438, 0.7031, 1.1406, 0.582, 1.3672, 0.8555, 1.1953, 1.0547, 0.9023, 0.7383, 1.0781, 1.5078, 0.7617, 1.5312, 0.7383, 0.8086, 1.5, 1.0938, 1.2344, 1.5781, 1.3672, 1.2266, 1.6172, 0.875, 1.3438, 1.6875, 1.2188, 1.1875, 0.832, 1.5, 1.0547, 1.3594, 0.7578, 0.9844, 0.9766, 1.2891, 1.3047, 1.1797, 1.1562, 1.0469, 1.3594, 0.8945, 1.2031, 1.75, 1.4297, 0.9531, 0.8945, 1.1328, 1.0547, 1.5234, 1.1406, 1.1328, 0.9219, 0.9648, 1.4062, 0.7656, 1.2734, 1.1484, 0.9766, 1.0703, 1.1562, 1.2812, 1.0781, 0.8789, 1.7734, 1.3359, 0.6211, 1.1641, 0.8633, 0.9492, 1.4609, 1.2188, 1.5391, 0.6133, 0.7617, 1.0469, 1.125, 1.5, 1.3438, 1.8984, 1.0781, 0.8945, 1.0938, 0.9219, 1.3203, 1.0625, 1.2422, 1.1562, 1.7344, 1.2266, 1.2344, 1.4141, 1.3672, 1.7266, 1.25, 1.3359, 1.1719, 0.7773, 1.3672, 1.8359, 1.0859, 1.2344, 0.6914, 1.1484, 1.4688, 1.0938, 0.8828, 1.3203, 1.0938, 1.2109, 1.3125, 1.1719, 1.4297, 1.1641, 1.2422, 1.2656, 1.2734, 0.9141, 1.2656, 1.4922, 1.6172, 0.9453, 0.9297, 1.0078, 1.1953, 1.125, 1.1719, 0.6602, 1.1484, 1.0469, 0.8242, 1.1328, 1.4688, 1.0234, 1.5859, 1.3516, 1.1094, 0.8477, 1.2656, 1.375, 1.2266, 1.6953, 0.9727, 0.8828, 0.6836, 1.0547, 1.6094, 1.4844, 1.5781, 1.2344, 1.0469, 0.7578, 1.2734, 0.9688, 0.9102, 0.8711, 0.7227, 1.3281, 1.2891, 1.4219, 1.1094, 1.6719, 1.1953, 1.4062, 1.1953, 1.1016, 1.2344, 1.3516, 0.6094, 1.3203, 0.9375, 0.5859, 1.2422, 1.0547, 0.7734, 1.3281, 1.0469, 0.9102, 1.6797, 1.3281, 1.0703, 1.2891, 0.8828, 1.4453, 0.7422, 0.9258, 0.9375, 1.2109, 0.7266, 1.1562, 1.6094, 0.9648, 1.3516, 1.0781, 1.1953, 1.4922, 1.0938, 0.7539, 1.4375, 0.9102, 1.0234, 1.375, 1.4297, 1.0156, 0.9531, 1.0625, 0.7852, 1.2969, 1.4062, 0.9531, 0.8438, 0.9336, 1.1406, 1.1875, 1.1719, 0.9922, 0.6328, 0.7891, 1.0, 1.5391, 1.5312, 0.8555, 0.957, 0.6719, 1.6328, 1.4297, 1.6562, 1.2344, 1.7031, 1.2188, 0.6094, 1.1016, 0.6992, 1.1797, 0.6562, 1.3359, 0.8594, 0.8828, 0.9062, 1.0156, 1.0078, 0.8828, 1.0312, 1.1875, 0.8242, 1.2031, 0.9023, 1.125, 1.5547, 1.0391, 0.8125, 1.2656, 1.0078, 1.6484, 1.2422, 1.3359, 1.4219, 1.5938, 0.9922, 0.9414, 0.8242, 1.3125, 1.1484, 1.3984, 0.6289, 1.3672, 1.0, 0.9258, 1.3281, 0.8711, 1.2031, 1.7266, 0.8711, 0.8086, 1.1016, 0.6719, 0.6133, 1.6406, 0.9844, 1.2188, 1.8906, 1.0234, 1.0547, 0.8633, 0.6719, 1.3203, 0.7969, 1.6094, 0.9492, 1.4531, 1.5312, 1.2578, 1.7578, 1.1094, 0.9141, 0.9375, 1.2891, 1.2344, 0.8789, 1.6641, 1.2031, 0.9336, 0.918, 1.3984, 1.4766, 0.9922, 1.2969, 1.0938, 0.7227, 0.75, 1.1875, 1.3984, 0.918, 1.3906, 1.1953, 1.1719, 1.2812, 1.125, 1.0625, 1.0781, 0.9883, 0.8516, 1.0938, 0.9922, 0.957, 0.7227, 0.793, 1.8047, 0.8516, 1.25, 1.25, 0.832, 1.0547, 1.5469, 0.8477, 1.3906, 0.9062, 1.0312, 0.7383, 1.6016, 0.8477, 1.3906, 0.9141, 1.1797, 0.7852, 1.1719, 1.0469, 1.0156, 0.9023, 1.3672, 1.0781, 1.0391, 1.0781, 1.1797, 1.1328, 1.2344, 1.1016, 1.3359, 1.2188, 0.7852, 0.7773, 0.7656, 1.8203, 0.9141, 1.0078, 1.4844, 1.2422, 0.7266, 1.0859, 1.4844, 1.0312, 1.5781, 0.9844, 1.1641, 1.5469, 1.25, 0.5742, 0.7227, 0.6094, 1.3125, 2.0312, 1.3203, 1.6484, 1.6172, 0.7695, 1.2031, 0.9648, 1.2266, 0.9453, 1.0, 0.7734, 1.3047, 1.3125, 1.0, 1.1328, 1.0938, 1.0, 1.1641, 1.0, 1.2344, 1.1875, 1.0234, 0.7773, 0.7773, 1.1953, 0.9805, 0.6797, 0.918, 0.9727, 1.5, 1.0391, 0.9688, 1.375, 1.2188, 1.0234, 1.1797, 1.0625, 0.6641, 1.0547, 0.832, 1.3672, 0.9297, 1.2109, 0.9844, 0.9609, 1.1016, 1.4375, 1.1094, 1.2188, 0.7695, 1.3594, 1.1172, 0.832, 0.8711, 0.9453, 0.707, 1.4062, 1.6484, 0.9531, 1.1016, 1.0703, 0.7812, 0.8789, 1.1641, 0.75, 0.6914, 0.7148, 1.0234, 1.5859, 1.4922, 1.4375, 0.9102, 1.4141, 1.3359, 1.1641, 1.3828, 1.1797, 1.4531, 1.5, 1.2578, 1.4062, 1.7031, 1.4688, 1.0234, 1.2109, 1.7266, 1.3906, 0.9375, 0.8555, 1.0703, 0.8555, 1.25, 1.7109, 0.9297, 1.1953, 1.4297, 0.9766, 1.3516, 1.5859, 0.5781, 1.2109, 1.2969, 1.0, 0.7695, 0.6562, 1.0391, 1.4531, 0.9844, 0.8203, 1.1641, 0.9336, 1.0156, 0.9766, 1.0469, 0.9688, 0.918, 1.1484, 1.1328, 1.4844, 1.1094, 1.1484, 1.9453, 1.2266, 1.875, 0.9297, 1.0859, 0.8164, 1.0078, 1.1719, 1.5547, 0.8125, 1.4453, 0.582, 1.3672, 1.4062, 1.1016, 1.4609, 0.8164, 1.0547, 0.793, 1.2266, 1.2969, 1.5156, 1.0859, 1.3594, 1.5469, 1.0938, 1.125, 1.6016, 1.1172, 0.7188, 1.1562, 0.9023, 1.1484, 0.875, 0.9375, 1.0156, 1.0469, 1.1562, 1.0547, 0.7773, 1.7109, 1.0234, 1.0156, 1.3984, 0.875, 1.0547, 1.3828, 1.0312, 0.8164, 1.1172, 1.0625, 0.9922, 1.2188, 1.25, 1.0547, 1.0938, 1.1484, 0.9727, 1.0625, 1.3438, 0.832, 1.2969, 1.0312, 1.4141, 1.4922, 1.4609, 1.4688, 0.6992, 1.4141, 1.0469, 1.0859, 0.9023, 0.6836, 1.125, 1.2031, 0.8828, 1.1094, 1.5312, 1.5469, 1.7109, 0.918, 1.1328, 1.4375, 1.4453, 1.5938, 1.5547, 1.3984, 1.3672, 1.4922, 1.3281, 1.8438, 0.9805, 1.0547, 1.5547, 0.9531, 0.7344, 2.4531, 1.3594, 1.1172, 1.0234, 1.0469, 1.125, 1.7031, 2.1094, 1.0703, 1.2891, 1.1094, 1.5781, 1.5703, 1.375, 0.9375, 1.2734, 1.0, 1.8047, 0.9219, 1.0312, 1.4688, 0.8906, 0.9844, 1.4219, 0.8555, 1.1406, 1.4453, 1.5547, 0.6836, 0.8555, 0.75, 1.3984, 0.8945, 1.3828, 1.125, 1.1172, 1.25, 0.9648, 1.0078, 1.5234, 1.3359, 1.0859, 1.0234, 0.6914, 1.2422, 0.9648, 1.4531, 1.3438, 0.7461, 1.4844, 1.0156, 1.2812, 1.7578, 0.9336, 1.4766, 0.8516, 1.1016, 1.5234, 0.9531, 0.8125, 0.8555, 1.4141, 0.7656, 1.0547, 1.3281, 1.4219, 1.0156, 1.4062, 1.1719, 0.9102, 1.25, 0.957, 0.8711, 0.9922, 0.9023, 0.9805, 0.9883, 1.8438, 1.0938, 0.9062, 0.6406, 0.7812, 1.0547, 0.7383, 1.1328, 1.1484, 1.3047, 1.1953, 0.918, 1.1094, 0.8828, 0.8555, 0.9609, 0.8047, 1.25, 1.3125, 1.2734, 0.9492, 0.7461, 1.0156, 0.9609, 1.5234, 1.0625, 1.9688, 1.1484, 1.0156, 1.25, 1.5391, 1.0938, 1.3906, 1.0156, 0.957, 1.0625, 1.2734, 1.4141, 1.0781, 1.25, 1.2031, 1.3672, 1.3125, 0.8672, 0.8594, 1.1875, 1.0391, 1.2188, 0.5742, 1.9453, 1.0859, 1.2188, 1.0938, 0.6992, 1.5156, 0.8164, 0.9062, 0.7734, 1.5312, 1.4531, 0.9531, 0.6523, 1.4766, 1.3359, 1.0625, 0.9453, 0.6953, 1.8516, 1.3984, 1.0, 1.0781, 1.1406, 0.9336, 1.3672, 0.7383, 1.1094, 1.5, 1.6641, 0.8594, 1.0547, 1.0781, 1.1406, 1.4609, 1.0781, 1.2656, 0.7734, 1.2109, 1.5, 1.0859, 0.9766, 1.3438, 1.25, 0.9688, 0.5977, 1.0391, 0.9844, 1.0781, 0.8359, 1.1406, 0.8867, 1.0547, 1.4688, 1.4062, 1.5547, 0.8867, 0.6211, 1.2344, 1.0, 1.375, 1.2188, 1.6641, 0.8008, 1.1094, 1.0312, 0.75, 0.6758, 1.6484, 1.2891, 1.0469, 1.8594, 0.7812, 1.8203, 1.5859, 1.1875, 1.5156, 1.3594, 1.25, 1.2266, 1.375, 1.0547, 1.4922, 1.0859, 1.0156, 0.9883, 1.3906, 1.6484, 0.8672, 2.1875, 0.7734, 0.9922, 1.5078, 0.8633, 0.8438, 1.0078, 0.7461, 1.2266, 1.125, 0.7227, 1.125, 0.7734, 0.7305, 0.8438, 1.0859, 1.1562, 1.5078, 0.7812, 1.2734, 1.4453, 1.2812, 0.9844, 1.6797, 1.4141, 1.0156, 0.5938, 1.3906, 0.7539, 1.2109, 1.9922, 0.9219, 1.2344, 1.375, 0.9453, 1.1953, 1.1953, 1.0391, 0.918, 0.8477, 0.9062, 0.8867, 0.8594, 1.1484, 1.25, 1.2812, 1.4688, 1.6406, 1.2188, 1.3047, 0.957, 1.0938, 1.5312, 1.0703, 1.1484, 1.0547, 0.9062, 1.0938, 1.4453, 1.3203, 1.2266, 0.6367, 0.8984, 1.4531, 1.6484, 1.1328, 0.8438, 1.8594, 0.7578]
|
Chapter 6/loss_comparison_attn_vlms.png
ADDED
|
Git LFS Details
|
Chapter 6/loss_comparison_attn_vlms_same_tokens.png
ADDED
|
Git LFS Details
|
Chapter 6/self_attention_batched_loss.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
[2.2656, 1.6328, 1.6016, 2.1406, 1.2578, 1.4375, 1.9141, 1.0234, 1.4922, 1.2969, 1.3906, 1.7812, 1.0703, 1.9297, 1.3516, 1.6719, 1.3203, 1.7266, 1.3359, 1.3516, 1.1328, 1.7109, 1.5234, 1.375, 2.2656, 1.625, 1.2422, 1.4375, 0.8125, 1.5859, 1.3672, 1.1953, 1.1016, 1.3984, 1.3047, 1.1797, 1.9844, 1.5781, 1.1875, 1.375, 1.7969, 0.7852, 1.4766, 1.3984, 1.4609, 1.2734, 1.7266, 1.3984, 2.3125, 1.3906, 1.3516, 1.5312, 1.0625, 1.3047, 1.7969, 1.6484, 1.1953, 1.4766, 1.7266, 1.6094, 1.7578, 1.7031, 1.2656, 1.9219, 1.1406, 1.1875, 1.4766, 1.5625, 1.5312, 1.3203, 0.9688, 1.6328, 1.0625, 1.25, 0.9375, 1.8672, 1.3984, 1.9609, 1.4531, 1.1641, 1.7578, 1.6016, 1.6953, 1.7031, 1.3203, 1.4531, 1.8125, 1.9062, 1.9453, 1.5625, 1.4141, 1.125, 1.1484, 1.2812, 1.6172, 1.7266, 1.1562, 1.7344, 1.3594, 1.0312, 1.2891, 1.3672, 1.125, 1.0547, 1.5469, 1.5312, 1.2422, 1.3438, 1.2656, 1.5625, 1.7812, 1.3203, 1.6172, 1.6875, 1.2422, 1.4844, 2.3125, 1.7734, 1.625, 1.1016, 1.1484, 1.2109, 1.3594, 1.6172, 1.6562, 1.3594, 1.0078, 1.4688, 1.0625, 1.6328, 1.2109, 1.4844, 1.1172, 1.125, 1.1328, 1.3906, 1.5156, 1.7188, 1.4688, 1.3672, 1.1797, 1.6406, 1.375, 1.2812, 1.875, 1.4375, 1.4297, 1.3203, 1.1641, 1.4609, 1.0, 1.0781, 1.7422, 1.4141, 0.9531, 1.5234, 2.125, 1.4922, 1.6328, 1.875, 1.1875, 1.1406, 0.9766, 1.7266, 1.1953, 1.2656, 1.0781, 1.6953, 1.6328, 1.5156, 1.4141, 1.9453, 0.957, 1.1797, 1.3516, 1.4844, 1.4766, 1.125, 1.9453, 1.2188, 1.1562, 1.3125, 1.5, 1.3125, 1.4219, 1.6016, 0.9883, 1.0469, 1.3047, 1.6172, 1.0625, 1.6016, 1.5938, 1.4375, 1.3672, 1.1094, 1.2656, 1.0234, 1.375, 1.6406, 1.1016, 1.2578, 1.3984, 1.1016, 1.8203, 1.5547, 1.3672, 1.2266, 1.0156, 1.3203, 1.3125, 1.5781, 1.6797, 1.4297, 1.1797, 1.4219, 1.2422, 2.4375, 1.2109, 1.625, 1.4219, 1.4688, 1.2109, 1.0312, 1.5469, 1.2578, 1.3203, 1.7891, 1.4062, 1.1641, 1.2109, 1.3672, 1.9141, 1.5391, 1.8281, 1.6562, 1.2266, 1.7344, 1.0391, 1.3672, 1.1328, 0.9922, 1.3906, 1.2031, 1.2344, 1.0156, 1.1719, 1.3672, 1.9609, 1.6641, 1.2266, 1.2891, 1.0781, 1.6406, 1.7891, 1.2109, 0.8867, 1.5859, 1.3828, 1.4141, 1.4766, 1.0156, 1.1328, 0.9297, 1.2188, 1.5781, 1.4688, 0.9336, 1.1172, 1.3281, 1.1016, 1.3203, 1.375, 0.8906, 1.5625, 1.6719, 1.6328, 1.1328, 1.4453, 1.4688, 0.8516, 0.8125, 1.2031, 1.4453, 1.0703, 1.1719, 1.1172, 1.3516, 1.3125, 1.3359, 1.0469, 1.0703, 0.9805, 1.8516, 1.0547, 1.6094, 1.0469, 1.25, 1.5234, 1.9219, 1.3047, 1.9531, 0.9609, 1.1406, 1.3906, 0.9805, 0.793, 1.5078, 1.5625, 1.3594, 1.6016, 1.2031, 1.3516, 1.4062, 1.2188, 1.5938, 0.9492, 1.4922, 1.6328, 1.2812, 1.3438, 1.0078, 1.625, 0.9297, 1.0859, 1.2578, 1.3203, 1.3438, 1.3047, 1.2891, 1.5391, 1.2578, 1.3281, 1.5859, 1.5625, 0.9727, 1.4297, 1.5859, 1.8516, 1.4844, 0.9102, 1.4141, 1.2969, 1.3203, 1.2344, 1.8906, 1.9609, 0.8867, 1.6328, 1.3281, 1.4844, 1.1562, 1.1953, 1.1641, 1.0469, 0.9219, 1.7266, 0.9766, 1.4766, 0.7578, 2.0781, 1.2656, 1.2422, 1.3906, 1.8594, 0.7734, 1.2969, 1.125, 1.2031, 1.3828, 1.3516, 1.0391, 1.0, 0.9609, 1.5938, 1.5312, 1.1953, 1.1406, 1.9219, 1.1016, 1.3984, 0.9531, 1.5156, 1.2422, 1.2656, 1.1562, 1.0469, 1.1094, 1.0859, 1.0703, 1.3203, 1.4375, 1.6172, 0.832, 1.2422, 1.3359, 1.3906, 1.0391, 1.4531, 1.0938, 1.6328, 1.75, 1.6875, 1.4922, 1.3359, 0.7969, 1.1484, 1.4688, 1.0234, 1.1406, 0.9219, 1.6484, 1.3594, 1.5469, 0.957, 1.8438, 1.2344, 0.8008, 1.4375, 1.2578, 0.7031, 1.4688, 1.3359, 1.2812, 1.2109, 1.3906, 1.0, 1.0781, 1.2109, 1.2578, 1.0781, 1.1875, 0.582, 1.2188, 1.2031, 1.3906, 1.0078, 1.3984, 1.1797, 1.5312, 1.1641, 1.7891, 1.3281, 1.7109, 1.0312, 2.4219, 1.5625, 1.0, 1.2188, 2.0469, 1.8203, 1.1641, 1.1016, 1.1719, 0.8086, 1.7734, 0.7891, 1.2188, 1.2031, 1.3594, 1.3906, 1.2344, 1.0547, 2.0469, 1.1562, 1.6953, 1.1953, 1.7109, 0.9727, 1.1641, 1.2188, 1.0156, 0.7188, 1.3828, 1.7188, 0.8945, 1.5625, 1.7031, 1.25, 1.4141, 1.1641, 0.8125, 1.7891, 1.2734, 1.0781, 1.125, 0.6914, 0.9961, 1.0703, 0.8438, 0.918, 0.8477, 2.0312, 1.1797, 0.8867, 2.2812, 0.9766, 0.7383, 1.0234, 1.2891, 1.6094, 1.4766, 1.5547, 1.1562, 1.6172, 1.0156, 1.4844, 0.918, 1.1328, 1.4688, 0.8711, 1.4766, 0.8555, 1.6484, 1.6641, 1.2188, 1.0859, 1.4375, 1.3125, 1.2422, 1.5312, 1.3047, 1.3047, 1.5625, 1.0312, 1.6094, 1.1875, 1.6328, 1.1328, 1.3984, 1.1719, 1.0781, 0.625, 1.3359, 1.25, 1.4062, 1.4922, 0.9648, 0.7383, 1.2266, 1.4062, 1.3125, 1.375, 1.0781, 1.5938, 0.6797, 1.4922, 0.8008, 1.4219, 0.918, 0.7891, 1.625, 1.2734, 1.4844, 1.1797, 1.9062, 0.7734, 1.7031, 1.1094, 1.0703, 1.3203, 1.6328, 0.9062, 1.1875, 1.2969, 1.8906, 0.8867, 1.4219, 1.2031, 1.4453, 1.4531, 1.3594, 1.7812, 1.4219, 1.0781, 1.1484, 1.5391, 1.4844, 1.5547, 2.0156, 1.4297, 0.8828, 1.125, 1.2578, 1.4375, 1.4062, 1.4609, 0.9531, 1.25, 1.2812, 1.5391, 1.2344, 1.5234, 1.4062, 1.1875, 1.3594, 1.2344, 1.2344, 1.2812, 1.0781, 1.3359, 1.4844, 1.5234, 1.1016, 1.0781, 1.1562, 0.7734, 1.4297, 1.4609, 1.5781, 1.2578, 1.0391, 2.0625, 1.3359, 1.3359, 1.7969, 1.4062, 0.8828, 1.4062, 1.3906, 1.1641, 1.5312, 1.2891, 1.0156, 1.4141, 1.1172, 1.3047, 1.1172, 1.2969, 1.3203, 0.9922, 1.8359, 1.4844, 1.2266, 1.2812, 1.5078, 1.25, 1.3281, 0.7773, 1.0, 1.2422, 1.3984, 1.5078, 1.1016, 1.1562, 1.3672, 0.875, 1.7891, 0.7148, 1.4297, 1.7422, 1.7031, 0.8164, 1.2422, 1.3203, 0.8633, 1.4531, 1.2578, 1.2891, 1.0547, 2.125, 1.2578, 1.7344, 1.4141, 1.0781, 1.0469, 1.0938, 1.1484, 1.3906, 1.1719, 1.4922, 0.8359, 1.0938, 1.0312, 1.1172, 1.3828, 1.1875, 1.6641, 0.9766, 1.5938, 1.1406, 1.3672, 1.0703, 1.3203, 1.1016, 1.5078, 0.7227, 1.7344, 0.8984, 1.4141, 1.3516, 1.5625, 0.6484, 1.5078, 1.0547, 1.4531, 1.2344, 1.1172, 0.6797, 1.4375, 1.2734, 1.6953, 2.1094, 1.1953, 1.2656, 1.2188, 1.3828, 1.2656, 1.6953, 0.8438, 1.4375, 1.3672, 1.5469, 1.25, 1.0, 1.2656, 1.4297, 1.4922, 1.4297, 1.5625, 1.2188, 0.9844, 1.0078, 1.7812, 1.7422, 1.2344, 1.4609, 1.625, 1.1172, 0.7773, 1.4531, 0.8672, 1.2109, 0.9453, 1.2891, 1.4375, 1.8203, 0.9102, 1.5234, 1.8281, 1.2031, 1.1562, 1.1953, 0.8594, 0.9648, 1.2969, 1.1562, 1.4375, 1.4766, 1.5547, 1.1406, 1.2031, 1.5469, 1.6641, 1.4141, 0.8359, 1.4453, 1.25, 1.6641, 1.3516, 0.8789, 0.9023, 1.5703, 1.2734, 1.4453, 1.3125, 1.1328, 1.3047, 0.8203, 1.4688, 0.9414, 1.2891, 1.2109, 0.8242, 1.1328, 1.25, 1.9375, 0.7852, 1.6016, 0.9102, 1.2812, 1.0234, 1.6094, 1.0469, 1.4609, 0.8672, 0.9297, 0.8125, 1.3125, 1.125, 1.8359, 1.3203, 1.2188, 1.6016, 0.668, 1.1875, 1.5547, 1.5938, 1.2266, 1.4062, 1.3125, 0.8711, 0.8164, 1.5469, 1.1719, 1.1641, 1.1484, 1.2266, 1.1641, 1.5781, 1.5234, 1.4922, 1.1562, 0.9102, 1.7422, 1.8203, 0.7969, 0.9922, 1.5703, 1.5156, 1.2578, 1.2422, 1.2578, 0.9805, 1.0391, 1.3984, 1.1719, 1.2031, 0.7188, 1.7344, 0.6445, 1.4531, 0.7031, 1.0703, 1.6719, 1.4766, 1.2656, 1.3906, 1.125, 1.2344, 0.75, 1.7891, 0.957, 1.2031, 1.7266, 1.9531, 1.2891, 1.7188, 1.4219, 1.6484, 0.9336, 1.4062, 1.2109, 1.1797, 0.8359, 1.0, 1.5, 1.0938, 1.2344, 1.4375, 1.4141, 1.3516, 1.6484, 1.1641, 0.7891, 0.7852, 1.3203, 0.9883, 1.25, 1.625, 1.0, 1.3906, 1.0312, 1.3516, 1.2969, 1.4062, 1.4609, 1.5859, 1.4219, 1.0859, 1.2891, 1.3984, 1.3359, 1.1328, 0.8711, 0.8516, 1.5859, 1.5234, 0.8398, 0.793, 0.9492, 1.6641, 1.1094, 1.1172, 1.3828, 1.2578, 1.3516, 1.5625, 1.1094, 1.3672, 1.7812, 0.9648, 1.2656, 1.6953, 0.9141, 1.5938, 0.9414, 1.1094, 1.0938, 0.6953, 0.9062, 0.8672, 1.5625, 1.1562, 0.8086, 0.7461, 1.8359, 0.7539, 1.7344, 1.6016, 1.2109, 1.125, 1.0781, 1.4141, 0.957, 1.8047, 1.5547, 1.0703, 1.6328, 0.9375, 1.0625, 1.2891, 0.9297, 0.9844, 0.8125, 1.4297, 1.2969, 1.2344, 1.2031, 1.5469, 1.2812, 1.6328, 0.8164, 1.1875, 1.1562, 1.1172, 1.4531, 1.1562, 1.2109, 0.7422, 1.6719, 0.9883, 1.5469, 1.2109, 0.7891, 0.7305, 1.3672, 0.9805, 1.3906, 1.2344, 1.6406, 1.6094, 1.0469, 1.3359, 0.9414, 1.3203, 1.2344, 1.1484, 1.3047, 0.8359, 1.2031, 1.375, 2.0156, 1.5781, 1.7812, 1.5859, 1.2812, 0.7148, 1.0781, 1.2188, 0.9414, 0.8164, 0.9414, 1.5938, 1.2656, 0.957, 1.0938, 1.1094, 1.2734, 1.5078, 1.6406, 1.1953, 0.9805, 1.7344, 0.9688, 1.3594, 1.2188, 1.2266, 1.4609, 1.4062, 1.1328, 1.6328, 1.5391, 1.2031, 0.8516, 1.0781, 1.0078, 0.7109, 1.8438, 0.8711, 0.7578, 1.0234, 0.9805, 1.6094, 1.1094, 0.9336, 1.2109, 1.7422, 1.2969, 0.7734, 1.6094, 1.3125, 1.2031, 1.0625, 1.0781, 1.5234, 1.1953, 1.3047, 0.9805, 0.75, 1.1484, 1.4609, 1.0469, 0.8594, 1.1484, 1.6328, 1.0703, 1.0625, 1.1953, 1.3359, 0.7773, 1.7266, 1.2344, 0.9297, 0.9062, 1.3125, 1.3359, 0.9414, 0.7344, 1.0391, 1.4531, 1.7578, 1.4062, 1.5547, 1.0156, 1.7422, 1.1328, 1.3672, 1.0547, 1.4062, 0.8477, 1.4453, 1.5156, 1.4688, 0.9492, 1.2734, 1.2031, 1.1719, 1.0703, 0.8828, 1.0312, 1.2891, 1.5391, 0.9766, 0.9297, 1.4922, 1.2109, 0.8867, 1.4453, 1.1406, 2.1406, 0.6953, 1.375, 0.8359, 2.0938, 1.3906, 0.9414, 1.4609, 1.1094, 1.1172, 1.2422, 1.3672, 0.7344, 1.1562, 0.8359, 0.8789, 0.9648, 0.8867, 0.8398, 1.1484, 0.9766, 1.4609, 1.0391, 1.0625, 1.3906, 1.7188, 1.2891, 0.8438, 1.125, 1.6719, 1.1797, 1.9531, 0.8594, 0.7734, 1.7031, 1.1641, 1.5703, 1.6406, 1.7109, 1.3438, 0.7969, 1.2188, 1.9141, 0.6914, 1.5469, 1.0, 1.2422, 1.0547, 1.4297, 1.5312, 1.8594, 1.2891, 0.9688, 0.9922, 0.9609, 0.8867, 0.8438, 1.4375, 1.2812, 1.2578, 1.3359, 1.0703, 0.9609, 1.1172, 1.5469, 1.2109, 1.0625, 1.4141, 1.0156, 1.3359, 0.9609, 0.9023, 1.4141, 1.0469, 0.8359, 1.6484, 1.3281, 1.2656, 0.8516, 1.7266, 1.2656, 1.1562, 1.1875, 1.8516, 0.9219, 1.3516, 1.1719, 1.0859, 1.3125, 1.4688, 0.8789, 0.9609, 1.0391, 1.6797, 1.2031, 1.6172, 0.8789, 1.0703, 1.2969, 1.2734, 1.1406, 1.5781, 1.3984, 1.1953, 0.9219, 1.4844, 1.7188, 1.0234, 1.3906, 1.2109, 0.8672, 0.7969, 1.0547, 1.0547, 1.2891, 0.9805, 0.9453, 1.1641, 1.3828, 1.1406, 0.8281, 1.3594, 1.0703, 1.1406, 1.9219, 1.5547, 1.5859, 0.9883, 0.8164, 1.1328, 1.5547, 1.2812, 0.9648, 1.8672, 1.0, 1.0234, 1.0703, 1.125, 1.1328, 1.3438, 1.3828, 1.1328, 1.0078, 1.1406, 1.7734, 1.1172, 0.8242, 0.957, 0.9531, 1.8516, 0.7383, 0.6875, 1.1875, 1.6172, 0.8672, 0.7734, 1.0078, 1.2109, 0.9258, 1.2109, 1.3906, 1.1953, 2.2188, 1.4531, 1.4062, 1.3828, 1.6484, 1.3906, 0.8906, 1.1641, 1.1016, 1.1328, 0.8125, 1.1016, 1.2188, 1.0156, 1.2656, 1.5, 1.0234, 1.125, 1.0703, 1.3047, 1.8984, 1.1016, 1.1562, 1.2578, 1.5234, 1.4609, 1.4531, 1.3516, 0.9492, 0.875, 1.1875, 1.3516, 1.0859, 1.2188, 1.5312, 1.5312, 0.8398, 0.8477, 0.8633, 1.3047, 1.3438, 1.125, 1.3359, 0.8203, 1.2734, 1.8438, 1.5234, 1.6953, 1.0703, 1.0781, 0.9297, 0.8086, 1.1719, 0.8555, 1.6875, 1.0, 0.7852, 1.5938, 1.2266, 1.1953, 0.9297, 1.2969, 1.4766, 1.3672, 1.1875, 1.2031, 1.4766, 0.9492, 1.4609, 1.5703, 1.1406, 0.832, 0.9414, 0.6367, 0.918, 1.8828, 0.9961, 1.5547, 1.5391, 1.0391, 1.0625, 1.3516, 0.8828, 1.1562, 1.2969, 1.1484, 1.0, 0.6758, 0.7852, 1.7969, 1.7109, 1.5391, 1.3906, 1.1641, 1.7734, 1.2812, 1.4141, 1.3828, 1.5, 1.0469, 1.3047, 1.6094, 0.8789, 1.5391, 1.5625, 1.5234, 0.8867, 1.1328, 0.9609, 1.1641, 1.4844, 0.75, 1.4297, 0.9727, 0.9922, 1.0547, 1.3906, 1.2266, 1.3203, 1.375, 1.1484, 1.4141, 1.3672, 1.4688, 1.5312, 1.3828, 0.7266, 1.3047, 1.3125, 1.2109, 1.4766, 1.2031, 1.75, 1.0703, 0.9297, 1.4766, 1.6094, 0.8516, 1.4375, 1.0156, 0.8242, 1.5234, 1.1172, 0.6562, 1.5391, 1.0, 1.3828, 1.4141, 1.3438, 1.4297, 1.7031, 1.3516, 0.9883, 1.5078, 1.1094, 1.0938, 1.6797, 1.6328, 1.3672, 1.4375, 0.9531, 1.1953, 1.1328, 1.7422, 1.3047, 1.8594, 1.1406, 1.0078, 1.2578, 1.25, 1.0859, 1.3438, 1.0938, 0.6758, 0.9805, 1.1719, 1.125, 1.7656, 1.3906, 0.9688, 1.3672, 1.3672, 1.4453, 1.1641, 1.2656, 1.0781, 0.5742, 1.1875, 1.25, 1.0547, 1.5234, 1.1406, 1.0703, 0.875, 1.3281, 0.9023, 1.5859, 0.8008, 1.0625, 1.4453, 1.1328, 1.3984, 0.9922, 0.9453, 0.7852, 1.3281, 1.2969, 1.2812, 1.3594, 1.3672, 1.1953, 1.5703, 1.1406, 1.7266, 1.1875, 1.2266, 1.2578, 1.0, 1.5078, 1.3906, 2.2656, 1.3516, 0.7539, 1.2969, 1.1719, 1.1719, 0.8242, 1.4531, 1.3125, 1.3438, 0.9102, 0.8047, 1.0703, 1.6641, 1.3594, 2.2344, 1.4141, 1.3672, 0.8438, 1.0781, 1.1484, 1.0938, 1.5625, 1.6016, 1.1953, 1.0703, 0.9883, 1.3906, 1.0391, 0.6602, 0.6562, 1.4531, 1.3516, 1.4375, 1.2656, 1.3594, 0.8398, 0.9297, 1.4219, 1.4844, 1.4141, 0.8984, 1.2266, 0.9688, 1.2891, 1.0469, 1.2891, 1.125, 1.0312, 1.7422, 0.9336, 1.2578, 1.3672, 1.3125, 1.3438, 1.6484, 1.0859, 1.4922, 1.4531, 1.2188, 1.1328, 1.1875, 1.1641, 1.4922, 1.4141, 1.4844, 1.2969, 1.5469, 1.0234, 1.5156, 0.7656, 1.6797, 1.1562, 1.2812, 1.3672, 1.1953, 1.2578, 1.4297, 1.7188, 1.3438, 0.9805, 0.9727, 1.25, 1.3203, 0.9062, 1.3516, 0.9609, 0.8438, 1.3281, 0.7031, 1.0, 1.4141, 1.4844, 0.9219, 1.4922, 1.6172, 0.918, 1.1953, 1.3984, 0.8984, 0.9766, 0.793, 1.3047, 1.3281, 1.2266, 1.4141, 0.8438, 1.2109, 0.9102, 1.5234, 1.375, 0.7148, 1.1172, 0.9609, 1.6953, 1.1484, 1.2266, 1.4453, 0.8672, 1.2812, 1.25, 1.3203, 1.4141, 1.6172, 1.1641, 0.9062, 0.8828, 0.7812, 1.2188, 0.9844, 1.6484, 1.0781, 0.918, 1.7812, 0.957, 0.9414, 1.1406, 1.4609, 0.6562, 1.4141, 1.4219, 1.5078, 1.0391, 1.1719, 0.7383, 1.6484, 1.3906, 2.0469, 0.9844, 1.3281, 0.6953, 1.1406, 1.2031, 1.3203, 1.2031, 0.6562, 0.9258, 1.1719, 0.7383, 0.9375, 1.3672, 1.1016, 1.1484, 1.9766, 1.0234, 1.5703, 1.6406, 1.2891, 1.1797, 1.0234, 0.7109, 1.1172, 0.9805, 1.1719, 1.7422, 1.1797, 1.5, 0.9531, 1.25, 1.0859, 1.1016, 1.1406, 1.2734, 1.1562, 1.0938, 1.6328, 0.8203, 1.3672, 0.8867, 1.4141, 1.0781, 1.2422, 0.9609, 0.9102, 1.2734, 1.125, 0.6641, 1.1641, 0.7422, 1.4609, 0.875, 1.0703, 1.1797, 1.5938, 0.9883, 0.9297, 1.5859, 1.3516, 1.1641, 1.1797, 1.0312, 1.1719, 0.75, 1.5625, 0.7422, 1.0312, 1.3828, 1.5703, 1.3516, 1.4375, 1.4141, 0.6562, 1.4531, 1.3438, 0.7969, 1.4062, 1.2656, 1.9219, 1.6172, 1.0859, 1.6562, 1.4375, 0.9531, 1.3047, 1.4531, 1.5234, 1.3281, 0.8555, 1.2891, 1.1406, 1.4453, 1.6484, 1.4766, 1.5781, 1.3594, 0.9023, 0.6406, 1.2344, 0.8164, 0.8477, 0.9102, 1.7344, 0.8867, 1.3203, 1.1406, 1.125, 1.0469, 1.3516, 0.8359, 1.1328, 1.2266, 1.4766, 1.3828, 0.7109, 1.2422, 1.6094, 1.1641, 1.1719, 1.2969, 1.0391, 1.125, 1.5078, 0.7656, 0.9023, 1.5156, 1.1172, 0.832, 1.5859, 1.3516, 1.2812, 0.9141, 1.3438, 1.7656, 1.0859, 0.9414, 0.8867, 0.9062, 0.8945, 1.2422, 0.9258, 1.0078, 1.0078, 1.6953, 1.6094, 1.3125, 2.125, 0.8398, 1.3047, 1.3438, 1.2031, 0.9297, 0.9492, 1.125, 1.8359, 0.625, 1.5078, 1.3828, 1.1172, 1.2188, 1.1641, 1.3125, 1.25, 1.4375, 1.2656, 1.2344, 1.375, 0.7891, 0.9688, 0.8281, 1.2266, 1.0938, 1.1719, 1.0938, 1.3672, 1.4844, 1.1562, 1.7891, 1.2891, 1.375, 1.4062, 1.4922, 0.793, 1.1094, 1.3438, 0.7422, 0.918, 1.2734, 0.9375, 1.2734, 0.6328, 0.9336, 1.5859, 0.6992, 1.2109, 1.1484, 1.2891, 1.0, 1.9062, 1.0234, 1.5938, 1.0234, 1.0859, 1.1562, 0.8906, 1.0469, 1.2266, 1.375, 1.1719, 1.0469, 1.5391, 1.4531, 0.9219, 1.2109, 1.2109, 1.2891, 1.2734, 0.7031, 0.7148, 1.2891, 1.3906, 0.8086, 1.4141, 0.9492, 1.1641, 1.2969, 1.2422, 0.9766, 1.3516, 1.0859, 1.0, 1.4453, 1.3359, 1.0391, 1.0078, 0.6836, 0.7617, 1.25, 0.8086, 1.3438, 1.6641, 1.0547, 1.1172, 0.957, 1.3203, 0.8242, 0.9609, 0.7539, 1.1719, 1.3594, 1.6406, 1.1016, 1.3594, 0.7734, 1.2188, 1.1641, 1.0938, 1.5859, 1.1875, 1.1172, 1.3359, 1.1094, 1.0156, 1.3125, 1.4609, 1.1875, 1.0234, 1.2812, 1.5, 1.5469, 1.4844, 0.8555, 1.1875, 1.1797, 1.4766, 1.3672, 1.0625, 1.1719, 1.0312, 1.4766, 2.4062, 0.9688, 1.0859, 1.0625, 0.8398, 1.1719, 1.1562, 1.3203, 1.4531, 1.0625, 1.0703, 1.4375, 1.3828, 0.9414, 1.2188, 1.4062, 1.2188, 0.9375, 1.1953, 1.0703, 1.7734, 1.0078, 0.8281, 1.125, 1.4453, 1.125, 1.5, 1.3984, 1.1484, 1.375, 1.125, 0.6641, 1.0938, 1.3047, 1.2734, 0.918, 1.2109, 1.1719, 0.957, 1.4219, 1.2422, 1.2578, 0.7461, 1.1719, 1.375, 0.8359, 0.9531, 1.0625, 1.6719, 1.6016, 1.4297, 1.4219, 1.375, 1.1328, 1.5547, 1.2969, 1.5938, 1.3125, 1.7109, 1.25, 1.6094, 1.0625, 0.8711, 1.0078, 1.3906, 0.957, 1.0078, 1.4375, 0.8867, 1.25, 0.7422, 1.2578, 1.6797, 1.1719, 1.0938, 1.6797, 1.1016, 0.8242, 1.4453, 1.2969, 1.2812, 1.0469, 1.1172, 1.0469, 1.5312, 1.0156, 1.1641, 1.2422, 0.8398, 1.4609, 0.7227, 1.1484, 1.3047, 0.9102, 1.7109, 1.1484, 1.5859, 1.2969, 1.3047, 1.6328, 0.8477, 1.2422, 1.5469, 1.1328, 1.2578, 0.9336, 1.1719, 1.1562, 1.2969, 1.6953, 0.8633, 1.0547, 1.1172, 1.2188, 1.1953, 1.4531, 1.0859, 0.8516, 1.2734, 1.6562, 2.0938, 1.3438, 0.8672, 1.1484, 1.0469, 0.8555, 1.1406, 1.4219, 1.7734, 0.6133, 1.3672, 1.375, 1.2031, 1.2109, 1.4688, 1.3438, 0.6133, 1.2266, 1.3906, 1.2031, 1.3516, 1.1562, 0.8633, 1.1875, 1.0859, 0.9688, 1.1953, 1.2266, 2.0781, 1.2266, 1.3984, 1.0078, 0.9883, 0.8789, 1.3125, 1.3438, 1.4609, 0.8945, 0.7852, 1.2422, 1.2422, 1.1016, 1.1875, 1.1641, 1.0703, 1.1016, 1.0547, 1.0703, 0.9102, 1.3828, 1.2188, 1.3438, 1.4219, 1.0078, 1.0312, 1.2109, 1.2344, 0.8438, 1.4297, 0.7422, 1.7891, 0.9414, 1.0625, 0.9102, 1.5078, 0.9453, 1.2344, 1.2188, 0.7695, 0.9531, 0.5039, 0.8047, 1.0391, 1.1953, 1.3828, 1.2812, 1.2578, 1.5156, 0.9844, 0.8633, 0.9453, 1.3828, 1.5234, 1.4453, 1.0234, 1.25, 1.2969, 1.5, 1.2344, 1.5938, 1.0156, 1.0703, 1.3281, 1.1172, 1.1328, 0.9688, 1.1797, 1.1953, 0.8555, 0.9102, 1.2344, 1.5156, 0.9922, 1.3906, 1.2344, 0.7852, 1.0312, 1.4219, 1.3828, 1.3672, 1.1094, 1.5781, 1.6016, 1.0781, 0.9062, 0.8633, 1.7578, 1.0703, 1.375, 0.9961, 1.4844, 0.7305, 1.1328, 0.8906, 1.0156, 1.25, 1.4297, 0.6055, 0.8828, 1.2812, 0.9336, 1.2969, 1.0312, 0.9102, 1.0703, 1.4141, 1.4688, 0.9922, 1.2969, 1.0, 0.8086, 1.2266, 1.4297, 1.1328, 0.625, 1.1641, 0.9375, 1.3594, 1.2188, 0.9375, 1.5781, 1.7344, 1.1172, 0.6641, 1.4766, 0.875, 1.3125, 1.0469, 1.3047, 0.8867, 1.0391, 0.9141, 1.1562, 1.0703, 1.5625, 1.0312, 0.8438, 1.2344, 1.4375, 1.0078, 0.9453, 1.5078, 1.4141, 1.5547, 1.0312, 1.2422, 1.0703, 1.5156, 1.1562, 1.1328, 1.1328, 1.0703, 0.7461, 1.2031, 0.7148, 0.8633, 0.9648, 1.2344, 0.7969, 1.0391, 0.6289, 0.9492, 0.8906, 1.7109, 0.6211, 1.3594, 1.4297, 1.125, 1.7344, 1.7812, 1.4219, 1.6172, 1.2578, 1.3203, 1.0156, 1.2734, 0.7539, 1.4062, 1.2188, 1.6719, 1.7188, 1.2031, 1.8203, 0.832, 1.0547, 1.2812, 1.2734, 1.6016, 1.2344, 1.3906, 1.1328, 1.5469, 1.2734, 1.3594, 1.125, 1.1406, 0.9922, 1.4297, 1.5391, 1.0, 1.0781, 0.957, 1.4062, 1.0234, 1.0938, 1.2188, 0.8594, 1.3125, 0.9258, 1.2109, 1.4062, 1.4688, 1.9453, 1.2031, 0.9336, 0.6289, 1.6172, 1.0078, 1.1094, 1.25, 1.1562, 0.6836, 1.3594, 1.0078, 0.7031, 0.9258, 1.2422, 0.6992, 1.5859, 1.0078, 0.707, 1.4844, 1.4297, 0.9258, 0.9883, 1.5156, 1.3828, 1.1953, 1.1875, 0.8164, 0.9883, 0.957, 1.6328, 1.3438, 0.957, 1.5469, 1.2109, 1.2891, 1.1875, 1.4141, 1.3594, 1.0781, 0.7188, 0.9453, 1.2188, 0.6914, 1.4766, 1.3359, 1.1719, 1.4375, 1.1328, 1.5469, 1.5469, 1.3906, 1.1016, 0.9961, 1.4141, 0.9961, 1.3438, 1.4844, 2.0312, 1.1953, 0.9297, 1.3984, 0.9297, 0.6367, 1.3281, 1.1484, 0.9258, 1.5547, 0.875, 0.8438, 0.9844, 1.7734, 1.1719, 0.832, 1.2266, 1.6094, 0.8242, 0.793, 1.7188, 1.5859, 1.2188, 1.4297, 1.0547, 1.4844, 0.793, 0.9844, 0.8398, 1.0547, 0.8516, 1.0703, 1.5625, 1.2812, 1.6172, 0.7266, 1.0703, 1.5625, 1.3906, 0.8516, 1.3125, 1.125, 1.5078, 0.9141, 0.8555, 1.0547, 0.8359, 0.9766, 1.4219, 1.2422, 1.7969, 1.2188, 1.1562, 1.3125, 1.2422, 1.25, 0.9023, 0.9141, 1.5312, 1.7344, 0.9844, 0.8477, 1.2344, 1.6641, 0.9414, 0.9961, 1.2109, 1.0, 1.3203, 1.2109, 1.0156, 1.6797, 0.9492, 1.5703, 0.9062, 2.125, 1.5703, 1.0859, 1.1953, 0.9531, 1.3359, 0.9375, 1.3047, 1.2422, 1.0078, 1.1875, 1.125, 0.7773, 1.1172, 1.5859, 1.0938, 1.0469, 1.3516, 1.0547, 1.3125, 1.1484, 1.2031, 0.6875, 1.0781, 1.2109, 1.3203, 0.8008, 1.6875, 1.25, 0.668, 1.3828, 1.4375, 0.8828, 1.3984, 1.125, 0.6914, 1.9375, 1.1562, 0.9297, 1.6719, 1.2656, 1.5547, 0.8789, 1.0781, 0.9336, 0.7656, 1.4141, 1.0781, 0.9648, 1.8594, 0.8047, 1.1875, 1.1875, 1.5234, 1.1641, 1.1172, 1.1094, 1.3438, 0.9453, 1.0938, 0.9258, 1.1016, 1.2812, 1.6875, 1.1562, 1.0312, 1.0625, 0.9453, 1.6406, 1.0625, 1.2734, 0.7617, 1.1484, 1.1406, 0.8203, 1.2109, 1.3828, 0.9844, 0.6172, 1.5781, 1.2578, 0.9023, 1.7188, 1.2734, 0.9375, 1.7656, 1.7969, 1.3281, 1.75, 1.3594, 0.8438, 1.1172, 0.9961, 1.2266, 1.0547, 1.4062, 0.9062, 1.25, 1.0938, 1.4922, 1.2266, 0.8945, 1.6328, 1.7656, 0.875, 1.2578, 1.1172, 0.7188, 1.0312, 1.1719, 1.6406, 1.2578, 1.3984, 1.4609, 1.1016, 0.8711, 1.4453, 1.6797, 1.0703, 0.9336, 1.0156, 1.1172, 0.9414, 0.8672, 1.4141, 1.2109, 1.0078, 0.8398, 1.2656, 1.5469, 0.9141, 1.3047, 1.0625, 0.8711, 1.7344, 1.2031, 0.918, 0.8281, 1.0078, 1.4766, 1.4766, 0.8711, 1.2344, 1.1328, 1.3438, 1.0078, 1.3516, 1.0156, 1.3125, 1.3828, 1.0391, 1.2969, 0.6719, 1.0234, 1.0547, 1.2578, 1.0469, 1.125, 1.0156, 1.3047, 1.1328, 1.4609, 0.6719, 1.7734, 1.3672, 1.3672, 1.4297, 1.0234, 1.2578, 0.7031, 1.4922, 1.0234, 1.0391, 1.5625, 1.1719, 1.2188, 1.3359, 1.0547, 1.4688, 1.0312, 1.0625, 1.5859, 1.2422, 1.6641, 1.2734, 1.3281, 1.0938, 1.3906, 1.3984, 0.9492, 1.4297, 0.8672, 1.1562, 1.1797, 0.7266, 1.5781, 1.2188, 1.5781, 1.3594, 1.6719, 1.5859, 1.4062, 1.7969, 1.1797, 0.793, 1.3203, 1.2656, 1.375, 0.9297, 1.7266, 1.1875, 0.9766, 1.125, 1.0859, 1.1484, 1.3281, 1.2188, 0.7148, 1.1094, 1.0469, 0.9531, 0.9336, 1.0, 1.2812, 1.7188, 0.9336, 0.9258, 1.1172, 1.6641, 1.4141, 0.9922, 1.0859, 1.3984, 0.8555, 1.2891, 1.4766, 0.8984, 1.1172, 1.0, 1.0781, 1.5625, 1.4062, 1.1797, 1.3438, 1.0625, 1.5703, 0.6836, 1.1719, 1.1484, 1.2344, 1.1562, 1.5938, 0.9336, 1.2578, 1.1719, 1.2422, 1.3203, 0.8906, 1.2188, 1.25, 0.8281, 0.9922, 1.0078, 1.8359, 1.2344, 0.8164, 1.125, 0.9375, 1.0312, 1.1094, 1.3828, 1.3359, 1.4609, 1.5938, 1.4609, 1.7812, 1.0234, 0.7266, 1.4375, 1.1484, 1.0859, 0.9883, 0.7539, 0.9805, 1.0, 0.6758, 0.7969, 1.0781, 0.8203, 0.7773, 0.9414, 1.5156, 1.2812, 1.4141, 0.8242, 1.5156, 0.6836, 0.6133, 0.9648, 1.5312, 1.125, 0.8867, 1.1719, 1.2109, 1.2734, 0.8359, 0.9453, 1.6328, 1.2188, 1.0781, 1.5469, 1.4453, 1.1797, 1.0469, 1.7109, 0.8828, 0.8594, 1.3047, 0.9297, 1.1641, 1.375, 0.8086, 1.1719, 0.9922, 1.4062, 1.125, 1.6797, 1.1406, 1.125, 1.3047, 1.2266, 1.1328, 1.1719, 1.5156, 1.1641, 1.1328, 1.2969, 1.4766, 1.3516, 0.7266, 0.6719, 1.3203, 1.3594, 1.1875, 1.2266, 1.0859, 1.2656, 1.4141, 1.0234, 0.8906, 1.4609, 0.8477, 1.2422, 1.1797, 1.1328, 1.0859, 1.3203, 0.8398, 0.7969, 0.9844, 1.1328, 1.3125, 1.3906, 1.0391, 0.9258, 1.2891, 1.5312, 1.3594, 1.3828, 1.5, 0.8281, 1.4609, 1.1094, 1.1953, 1.0547, 1.375, 1.1094, 0.9844, 0.7695, 1.5, 1.5469, 0.9922, 1.2266, 0.8477, 1.1328, 1.1641, 0.9375, 1.1641, 1.4766, 1.0312, 1.1328, 1.5078, 1.6719, 1.4531, 1.4922, 1.0859, 1.9766, 1.3984, 1.4844, 1.3359, 1.3047, 1.3203, 1.1641, 0.8125, 1.1719, 1.0312, 1.2969, 1.0547, 1.2578, 1.2656, 0.957, 0.7461, 0.8008, 1.4141, 1.1016, 0.7461, 1.4922, 0.793, 1.2031, 1.2578, 1.125, 1.0, 0.8164, 1.1406, 0.9375, 1.2188, 1.2266, 1.0547, 1.25, 0.9727, 1.2969, 1.0312, 0.9023, 1.5234, 1.4141, 1.8438, 1.4688, 1.3828, 1.4062, 1.1172, 1.0078, 1.0391, 1.2891, 1.4219, 0.8047, 1.6328, 1.0625, 1.1562, 0.6484, 0.7305, 0.9609, 1.5312, 1.1719, 0.9531, 1.0312, 1.1484, 0.9297, 1.0547, 1.5312, 1.1172, 1.3516, 0.8281, 0.9297, 0.8633, 1.3281, 1.6875, 1.0625, 1.2344, 1.3984, 0.9531, 1.1172, 1.1094, 1.1875, 0.9648, 1.2578, 1.1328, 1.0938, 1.4375, 0.918, 0.6211, 1.375, 1.2656, 0.9922, 0.8203, 1.0781, 0.9883, 0.7891, 1.1172, 1.0391, 1.0078, 1.5625, 0.9766, 1.3984, 0.9648, 1.0781, 1.1797, 1.0312, 1.25, 1.6172, 2.2812, 1.2188, 0.8477, 1.1328, 1.4922, 1.6016, 1.0156, 1.6953, 1.0234, 1.6172, 1.0781, 0.9297, 1.3281, 1.0469, 1.0781, 1.3047, 0.9102, 1.5, 1.5781, 0.9102, 1.2891, 0.7031, 0.9375, 0.6719, 1.1875, 0.7148, 0.8984, 0.6289, 1.6641, 1.2578, 1.4609, 1.0703, 1.0391, 1.4766, 1.1875, 0.9609, 0.9844, 0.9062, 0.8086, 0.8906, 1.0078, 1.6719, 1.6094, 1.2266, 1.125, 1.6406, 1.4219, 0.7656, 0.957, 1.7969, 1.0859, 1.0859, 1.5234, 1.0312, 1.0234, 1.0156, 1.1719, 0.9688, 1.1484, 1.1875, 0.9609, 1.0625, 1.0938, 1.0781, 1.5, 1.1641, 1.5312, 1.0078, 0.9023, 1.1406, 1.1562, 1.0, 1.3438, 1.2266, 0.7539, 0.8516, 0.8867, 0.793, 0.8906, 0.8945, 0.8281, 1.0781, 1.4531, 0.8203, 1.5234, 1.4375, 1.0859, 1.1094, 1.0234, 0.6953, 1.1562, 1.3516, 1.1562, 0.7969, 1.0859, 1.5078, 0.9297, 1.3984, 1.3438, 1.5781, 1.1719, 1.0078, 0.9531, 1.125, 1.0859, 1.2188, 1.3203, 1.1875, 1.4219, 1.6094, 1.75, 1.2812, 0.8945, 0.7461, 1.1172, 1.5234, 1.3438, 0.8633, 0.9883, 1.8516, 1.0156, 1.0, 1.1719, 1.1641, 1.6172, 1.3047, 0.9766, 1.3516, 0.707, 1.3594, 0.9922, 1.2734, 0.9336, 1.1484, 1.4141, 0.9883, 1.7109, 1.5547, 0.6875, 0.8828, 1.1328, 1.2969, 0.9492, 0.8125, 1.1641, 1.3047, 0.9805, 1.125, 1.3125, 1.0781, 0.9727, 0.707, 1.1016, 1.1797, 1.3438, 1.0547, 1.4453, 1.2891, 0.5508, 0.6953, 1.1172, 1.1641, 1.3672, 1.0938, 1.3203, 0.7617, 1.0625, 1.2656, 0.8633, 0.8047, 1.1094, 1.2188, 0.9883, 1.6719, 1.4766, 0.8711, 1.3438, 1.3672, 0.8867, 1.6328, 1.2109, 0.9023, 1.4609, 1.2109, 0.875, 1.3594, 1.3047, 0.8164, 1.5312, 0.9297, 1.4062, 1.2891, 1.1016, 0.7266, 1.0391, 1.3828, 1.2656, 1.1484, 0.8164, 0.8242, 1.1719, 0.7031, 0.9492, 0.9297, 1.2266, 0.957, 1.1328, 1.2109, 1.2734, 1.5625, 0.8945, 1.7891, 0.9922, 1.0156, 1.2344, 0.9336, 1.4531, 1.9297, 1.2031, 0.9297, 1.4219, 0.9648, 1.3047, 1.1953, 1.2188, 0.7148, 1.2969, 1.5391, 0.8047, 0.875, 1.2031, 1.1328, 1.4219, 1.4453, 1.0703, 1.0, 1.2422, 1.5625, 1.4531, 1.6719, 1.0938, 0.9102, 1.5703, 0.8164, 1.2656, 0.8242, 1.2422, 1.6562, 1.4766, 0.7539, 1.125, 1.0547, 1.3359, 1.1719, 1.2422, 2.0781, 1.0547, 1.6797, 1.2656, 1.1328, 0.7344, 1.0859, 0.5859, 1.1875, 1.3281, 1.1328, 1.4453, 1.6484, 0.7812, 1.6016, 1.2578, 1.3203, 1.25, 1.2188, 0.8672, 1.5234, 1.2031, 1.1719, 1.0469, 0.75, 1.8828, 1.5234, 0.832, 1.2656, 1.0078, 1.0234, 0.7266, 1.0625, 1.25, 1.3906, 0.832, 0.6758, 1.7734, 1.4609, 0.9883, 0.9805, 1.1172, 1.4531, 1.1328, 1.9141, 1.7578, 0.9648, 1.3828, 1.0703, 0.8828, 1.2188, 1.2969, 0.8242, 1.1016, 0.9805, 0.9336, 1.4844, 0.7344, 0.9414, 1.8906, 1.3516, 1.0078, 0.8008, 0.9609, 1.3984, 0.7695, 1.3828, 1.5078, 1.2656, 1.0859, 1.5, 1.4297, 1.9297, 1.1406, 1.3438, 1.1406, 0.9531, 1.1172, 1.4297, 0.9961, 1.2422, 0.9023, 1.2734, 1.0469, 0.8594, 1.0156, 1.3828, 1.7891, 1.2344, 1.5156, 1.4219, 0.9102, 1.75, 0.8438, 1.2578, 1.4844, 0.9766, 1.0, 1.2266, 1.2344, 1.3203, 0.7656, 2.5156, 0.6953, 1.6641, 1.0625, 1.6562, 1.0, 0.7852, 1.0703, 0.9922, 1.3047, 0.8984, 1.2578, 0.9805, 1.1406, 0.7734, 1.3906, 1.3125, 1.125, 1.0312, 0.5547, 1.2656, 1.2109, 0.9336, 1.0312, 1.4062, 0.9336, 0.8984, 1.2109, 1.4609, 1.1797, 1.0312, 1.4375, 0.832, 1.1719, 0.9531, 1.0391, 1.0469, 1.2188, 0.8867, 1.3906, 1.5391, 1.0938, 0.6289, 0.7891, 0.9492, 0.5312, 1.0781, 0.8242, 0.9023, 1.25, 1.3281, 1.3438, 1.3516, 1.2422, 0.7461, 1.0391, 1.5234, 1.1641, 0.7148, 1.1953, 1.1172, 1.3594, 1.1094, 1.1406, 1.1719, 1.2578, 1.0703, 0.8164, 1.5859, 1.2422, 0.8086, 1.0234, 1.4922, 1.1797, 0.9297, 0.9844, 1.0625, 1.0469, 0.9961, 1.2812, 1.0703, 2.0312, 0.7891, 0.9453, 1.1797, 1.0312, 1.0078, 0.7539, 1.125, 0.7422, 1.1172, 1.2891, 1.2656, 1.5781, 0.8984, 0.9844, 1.0156, 1.2812, 0.8281, 1.75, 0.9727, 1.0, 1.1484, 0.9688, 1.1406, 1.1328, 1.4766, 1.0, 1.4219, 1.0391, 1.3125, 1.0391, 1.0781, 1.25, 1.3984, 1.9609, 1.4062, 1.5156, 1.2812, 1.1172, 1.1172, 1.3516, 0.9414, 1.3125, 0.957, 0.6562, 1.1094, 0.9219, 1.6016, 1.3438, 1.0078, 1.1719, 1.5, 1.1406, 1.7422, 1.1172, 0.8984, 1.0625, 0.6758, 0.8867, 1.0469, 0.582, 0.9492, 1.2891, 0.8164, 1.6562, 0.9609, 0.832, 1.2812, 1.2109, 1.4297, 1.3672, 1.2109, 1.0547, 1.1875, 1.0938, 1.8359, 1.5625, 1.1484, 1.0156, 1.375, 1.2422, 1.1094, 1.6953, 1.1641, 1.0859, 1.1641, 1.0703, 0.9727, 1.1719, 1.4297, 0.9219, 1.2656, 1.2891, 1.1328, 0.8828, 0.9844, 0.8789, 1.0547, 1.2734, 2.0312, 1.4141, 1.1562, 1.4688, 1.125, 1.1562, 1.1172, 1.3203, 1.2031, 0.8906, 0.7148, 0.918, 0.957, 1.1719, 1.5469, 1.3672, 1.2109, 1.3516, 0.8281, 1.3594, 1.0156, 1.1562, 1.2656, 1.0547, 1.0156, 0.6797, 1.4688, 1.375, 1.5781, 0.9883, 1.4453, 0.9414, 0.7383, 1.3125, 1.6406, 1.4062, 0.7773, 1.0625, 0.6992, 0.7422, 1.2734, 1.0234, 1.0781, 1.2656, 0.9492, 1.7109, 1.0938, 1.0156, 1.7422, 1.2812, 1.3828, 1.375, 0.6719, 1.4219, 0.7812, 0.9727, 0.7656, 0.707, 1.1797, 1.8047, 1.1875, 1.2812, 1.2031, 0.7344, 0.957, 0.8555, 1.6406, 1.1016, 1.0312, 1.9766, 1.0, 1.0859, 1.1719, 1.0938, 1.7266, 1.1016, 1.5859, 1.4219, 0.8633, 1.2422, 0.9922, 1.1641, 1.9062, 1.0078, 1.1328, 1.0469, 0.5859, 1.1406, 1.0078, 0.5664, 1.0156, 0.8086, 1.6719, 0.9141, 1.1953, 1.1719, 1.5156, 1.3516, 0.9727, 1.4609, 1.1406, 1.2734, 0.8047, 1.4219, 0.8398, 0.75, 1.75, 0.8711, 0.9492, 1.2031, 1.1328, 0.8203, 0.6875, 1.7422, 0.7617, 0.8906, 1.0078, 1.2344, 1.1953, 1.2188, 1.0, 0.9336, 1.2344, 1.2891, 0.8281, 1.2031, 1.0156, 1.1641, 0.7227, 0.7969, 1.0, 1.1016, 1.1172, 0.875, 1.5547, 0.8125, 1.1094, 0.9492, 1.4219, 1.4375, 1.8516, 1.1562, 1.3906, 1.0703, 1.1641, 1.0625, 1.2344, 1.3047, 1.1172, 1.2422, 1.3047, 1.6875, 1.4453, 0.7773, 1.5859, 0.7305, 1.1016, 1.4219, 1.0781, 0.7305, 1.3672, 1.3438, 1.2266, 1.0703, 1.2734, 1.5547, 1.125, 1.5469, 1.25, 1.0625, 0.9766, 0.875, 1.3281, 1.3828, 1.0312, 1.0859, 1.1562, 1.3594, 1.0859, 1.4766, 1.2891, 1.0625, 0.8984, 1.2812, 0.8594, 1.1953, 0.8711, 1.1641, 1.0703, 1.0, 1.3125, 0.9297, 0.7734, 1.3984, 1.3203, 1.0469, 1.3516, 1.125, 0.8164, 0.9688, 0.9258, 0.9883, 0.7305, 0.832, 1.2734, 1.3828, 0.7695, 0.8086, 0.7266, 0.9531, 1.2422, 0.8359, 0.8633, 1.3672, 1.0938, 1.1562, 1.0078, 0.9453, 1.0, 1.0469, 1.3438, 1.2188, 1.2031, 1.75, 1.0234, 1.0469, 0.6875, 1.1328, 0.9453, 1.2578, 0.8438, 1.5469, 1.8281, 1.5234, 0.9375, 1.4141, 1.5078, 0.9688, 0.9375, 1.6328, 0.8945, 0.8672, 1.7266, 1.1875, 0.7852, 1.6172, 0.7227, 1.7891, 1.3828, 0.8984, 0.9414, 0.9258, 0.8945, 1.2734, 0.7422, 1.0156, 0.9375, 1.1094, 1.2109, 0.9492, 1.5781, 1.0625, 0.7266, 1.1797, 0.9609, 1.3906, 1.3438, 1.4609, 0.9766, 1.2656, 1.1484, 1.2812, 1.7344, 0.9062, 1.0703, 1.4453, 1.5938, 1.1641, 1.2891, 1.0156, 1.5625, 0.7656, 1.7812, 0.9805, 0.8555, 0.8984, 1.1953, 1.0547, 1.2969, 1.3594, 0.9336, 1.4922, 0.8359, 0.8672, 1.2578, 1.1328, 1.4219, 0.9414, 1.6797, 0.8359, 0.9141, 0.9844, 0.9648, 1.6172, 1.2812, 0.9883, 1.3281, 1.2969, 1.2109, 1.0859, 1.5859, 1.6328, 0.793, 1.1797, 1.2188, 0.9219, 0.6523, 1.1641, 1.4375, 0.8945, 0.9102, 1.3594, 1.3203, 0.9648, 1.3203, 1.0781, 0.668, 1.2266, 1.1641, 1.1797, 1.3438, 1.0859, 1.3594, 1.2656, 1.4531, 0.9805, 0.9375, 0.875, 1.2656, 1.0469, 1.3125, 0.9961, 1.6641, 0.9219, 1.3281, 1.3516, 0.8398, 0.8281, 1.2422, 1.3672, 1.25, 1.3359, 1.0547, 1.3672, 1.1797, 1.3672, 1.2812, 1.2969, 1.7812, 0.8438, 0.9375, 1.2891, 1.6406, 0.7734, 1.0391, 1.0781, 1.2578, 1.1094, 1.0547, 1.1641, 1.1406, 0.8438, 1.3906, 1.2188, 0.9883, 1.5547, 1.4375, 1.7656, 1.4219, 1.3984, 1.1797, 0.9688, 0.7461, 0.8047, 1.2969, 0.7969, 1.0469, 1.0469, 1.0234, 0.7188, 0.707, 1.0625, 0.8789, 1.3125, 1.0625, 1.0781, 1.4688, 1.2266, 1.0547, 1.2109, 1.6875, 0.7109, 1.5, 1.5703, 0.6836, 1.4922, 1.3594, 1.2344, 1.0781, 1.1719, 0.7695, 1.2109, 1.2891, 1.2344, 1.3516, 1.0156, 1.2969, 1.2031, 0.7578, 1.0078, 1.1641, 0.7578, 1.0078, 1.1719, 1.1172, 1.3359, 1.0391, 1.3594, 1.9375, 0.9492, 0.9297, 0.6641, 0.9453, 1.0234, 1.0859, 1.0156, 0.9453, 1.3359, 1.3984, 1.0, 0.7891, 0.832, 1.9688, 1.1719, 0.9297, 1.1953, 1.1406, 1.2891, 1.1484, 0.7148, 1.7109, 1.5781, 1.7422, 1.0547, 1.6797, 0.8789, 0.9375, 0.9062, 1.2109, 1.5859, 0.9492, 0.8555, 0.8516, 1.1484, 1.4297, 1.2266, 1.2656, 0.7891, 1.3047, 0.9375, 1.1562, 0.7969, 1.5, 0.8008, 0.8516, 0.9648, 1.2344, 0.9648, 1.3438, 1.3438, 1.2578, 1.1016, 1.4297, 1.2891, 0.8477, 0.8008, 1.7109, 1.7344, 1.4141, 1.1953, 1.5234, 1.5938, 1.4141, 0.9727, 0.7344, 1.125, 0.5547, 1.2891, 0.8086, 1.1641, 1.1719, 1.3125, 1.5547, 0.9336, 1.4609, 1.4297, 1.2812, 1.5703, 1.125, 1.3047, 1.2812, 0.9258, 0.6602, 1.3438, 1.0156, 1.3438, 1.4062, 1.3438, 0.9609, 1.2969, 0.7969, 1.2812, 1.3594, 0.9453, 1.0, 0.9492, 1.4141, 0.9727, 1.8203, 0.832, 1.1953, 1.4844, 1.1562, 1.3125, 1.1875, 0.7227, 1.3281, 1.1797, 1.4062, 0.9883, 0.7969, 1.4297, 0.8008, 0.6367, 1.6719, 0.9453, 0.7266, 2.0781, 1.2344, 1.5781, 1.1797, 1.4844, 1.1875, 1.0859, 1.0078, 1.2266, 1.2109, 1.375, 1.3438, 1.4375, 1.0156, 1.6562, 0.9531, 1.5469, 0.7344, 0.9414, 1.1328, 1.0859, 0.9688, 1.125, 1.4844, 1.4531, 1.0703, 1.0547, 0.9922, 1.2266, 1.125, 1.4297, 1.7656, 0.832, 0.9805, 1.4688, 1.1094, 1.4688, 0.918, 1.1484, 0.9648, 0.918, 0.9648, 0.7031, 0.9453, 0.6914, 1.1406, 0.8047, 0.8555, 1.5312, 1.3984, 1.1641, 0.7578, 1.1875, 0.9297, 0.9688, 1.0078, 1.2109, 0.6719, 1.1641, 0.9102, 0.8164, 1.0859, 1.1328, 1.5, 0.875, 1.2734, 1.3516, 1.1172, 1.4922, 1.7891, 1.5, 0.8555, 1.0938, 1.4844, 0.5898, 1.6719, 0.8555, 1.2891, 1.2109, 1.3984, 0.7344, 1.2578, 1.3906, 1.4688, 0.9492, 1.1016, 1.4062, 0.9414, 0.957, 0.8594, 1.0625, 0.9414, 1.8672, 0.7461, 1.2812, 0.918, 1.2344, 0.9258, 1.4297, 1.7344, 1.2578, 0.7852, 1.2031, 1.0469, 0.8047, 1.7734, 0.7969, 1.0938, 0.9766, 0.8672, 1.3672, 0.9336, 1.2969, 0.9609, 1.6875, 1.6562, 0.7109, 1.3984, 1.3125, 1.0703, 1.8438, 1.0078, 1.3438, 1.2188, 0.5273, 0.8203, 1.3516, 1.3672, 1.4297, 1.2422, 1.1641, 1.1172, 0.9805, 2.1562, 0.9297, 0.7773, 0.8984, 0.8672, 1.6406, 0.9062, 1.0703, 1.4062, 1.3984, 0.9609, 1.4297, 0.6562, 1.1016, 0.9219, 0.7617, 1.6172, 0.8438, 1.25, 1.3516, 0.7109, 1.1484, 0.5859, 1.375, 0.8594, 1.2031, 1.0625, 0.9141, 0.7422, 1.0859, 1.5078, 0.7734, 1.5469, 0.7539, 0.8164, 1.5, 1.0938, 1.2344, 1.5938, 1.375, 1.2422, 1.6094, 0.875, 1.3438, 1.6953, 1.2188, 1.1953, 0.8398, 1.5078, 1.0547, 1.3672, 0.7773, 0.9961, 0.9844, 1.2969, 1.3125, 1.1797, 1.1719, 1.0469, 1.3672, 0.9141, 1.2031, 1.75, 1.4375, 0.9414, 0.9141, 1.1328, 1.0703, 1.5391, 1.1484, 1.1484, 0.9297, 0.9727, 1.4062, 0.7773, 1.2812, 1.1562, 0.9844, 1.0703, 1.1719, 1.2969, 1.0938, 0.8867, 1.7812, 1.3438, 0.625, 1.1719, 0.875, 0.9531, 1.4766, 1.2344, 1.5547, 0.6133, 0.7695, 1.0547, 1.1406, 1.5, 1.3359, 1.9062, 1.0781, 0.9102, 1.0938, 0.918, 1.3125, 1.0625, 1.25, 1.1641, 1.7344, 1.2422, 1.2422, 1.4297, 1.3906, 1.7422, 1.2734, 1.3516, 1.1719, 0.7969, 1.3906, 1.8281, 1.1094, 1.2344, 0.6992, 1.1484, 1.4844, 1.0938, 0.8867, 1.3203, 1.1016, 1.2188, 1.3125, 1.1719, 1.4453, 1.1641, 1.25, 1.2656, 1.2734, 0.9258, 1.2812, 1.5078, 1.625, 0.9492, 0.9375, 1.0078, 1.2188, 1.1328, 1.1719, 0.668, 1.1641, 1.0547, 0.8242, 1.1406, 1.4766, 1.0391, 1.5859, 1.3594, 1.1094, 0.8516, 1.2734, 1.3672, 1.2344, 1.6953, 0.9805, 0.8867, 0.6836, 1.0469, 1.625, 1.4844, 1.5859, 1.2422, 1.0625, 0.7617, 1.2656, 0.9727, 0.918, 0.8789, 0.7227, 1.3359, 1.2969, 1.4219, 1.1094, 1.6641, 1.2109, 1.3984, 1.1953, 1.1016, 1.2422, 1.3594, 0.6133, 1.3281, 0.9453, 0.5977, 1.25, 1.0625, 0.7773, 1.3359, 1.0547, 0.9219, 1.6797, 1.3281, 1.0781, 1.2891, 0.8867, 1.4531, 0.7422, 0.9336, 0.9414, 1.2109, 0.7383, 1.1719, 1.6172, 0.9688, 1.3438, 1.0859, 1.1953, 1.5078, 1.1094, 0.7578, 1.4453, 0.9141, 1.0234, 1.3906, 1.4375, 1.0156, 0.9609, 1.0547, 0.7852, 1.2969, 1.4062, 0.9531, 0.8477, 0.9336, 1.1484, 1.1875, 1.1719, 1.0078, 0.6367, 0.7969, 1.0078, 1.5469, 1.5234, 0.875, 0.9648, 0.6758, 1.6406, 1.4297, 1.6641, 1.2344, 1.7188, 1.2266, 0.6133, 1.1016, 0.7109, 1.1875, 0.6602, 1.3359, 0.8672, 0.8906, 0.9062, 1.0234, 1.0312, 0.8828, 1.0312, 1.1875, 0.832, 1.2109, 0.9023, 1.1406, 1.5625, 1.0469, 0.8164, 1.2734, 1.0156, 1.6562, 1.2578, 1.3359, 1.4219, 1.6094, 0.9922, 0.9375, 0.832, 1.3203, 1.1641, 1.3984, 0.6367, 1.3828, 1.0, 0.9414, 1.3281, 0.8711, 1.2188, 1.7344, 0.875, 0.8086, 1.1094, 0.6797, 0.625, 1.6484, 0.9883, 1.2266, 1.8906, 1.0234, 1.0547, 0.8672, 0.6797, 1.3203, 0.8008, 1.6172, 0.9492, 1.4609, 1.5312, 1.2656, 1.7734, 1.1172, 0.9219, 0.9609, 1.3047, 1.2422, 0.8828, 1.6641, 1.2031, 0.9453, 0.918, 1.4062, 1.4844, 1.0, 1.3047, 1.0938, 0.7305, 0.7578, 1.2031, 1.4062, 0.9219, 1.3984, 1.1953, 1.1875, 1.2891, 1.125, 1.0625, 1.0938, 0.9805, 0.8594, 1.1094, 0.9961, 0.9648, 0.7227, 0.7969, 1.8125, 0.8594, 1.2734, 1.2578, 0.832, 1.0547, 1.5469, 0.8711, 1.3984, 0.9102, 1.0469, 0.7461, 1.6094, 0.8555, 1.4219, 0.9219, 1.1797, 0.8047, 1.1719, 1.0625, 1.0234, 0.9023, 1.375, 1.0859, 1.0547, 1.0859, 1.1953, 1.1406, 1.2422, 1.1016, 1.3359, 1.2266, 0.7891, 0.7852, 0.7734, 1.8438, 0.9219, 1.0234, 1.4844, 1.25, 0.7305, 1.1016, 1.4844, 1.0547, 1.5781, 0.9922, 1.1719, 1.5547, 1.25, 0.5781, 0.7266, 0.6289, 1.3281, 2.0469, 1.3125, 1.6484, 1.6172, 0.7695, 1.2188, 0.9766, 1.2266, 0.9609, 1.0078, 0.7812, 1.3125, 1.3281, 1.0, 1.1406, 1.1016, 1.0078, 1.1797, 1.0078, 1.2344, 1.2109, 1.0312, 0.7773, 0.7773, 1.2031, 0.9844, 0.6836, 0.9297, 0.9805, 1.5156, 1.0469, 0.9688, 1.3828, 1.2188, 1.0234, 1.1953, 1.0703, 0.6602, 1.0469, 0.8555, 1.375, 0.9258, 1.2188, 0.9961, 0.9648, 1.1172, 1.4609, 1.125, 1.25, 0.7617, 1.3594, 1.1406, 0.8359, 0.8711, 0.9492, 0.707, 1.4219, 1.6562, 0.9648, 1.1172, 1.0859, 0.7969, 0.8828, 1.1719, 0.7656, 0.6953, 0.7148, 1.0234, 1.5938, 1.4922, 1.4609, 0.9258, 1.4297, 1.3438, 1.1562, 1.3984, 1.1797, 1.4531, 1.5156, 1.2734, 1.4062, 1.7031, 1.4766, 1.0234, 1.2188, 1.7266, 1.3906, 0.9375, 0.8516, 1.0781, 0.875, 1.2578, 1.7109, 0.9375, 1.2031, 1.4453, 0.9766, 1.3516, 1.6016, 0.5859, 1.2266, 1.3047, 1.0078, 0.7773, 0.668, 1.0469, 1.4609, 0.9961, 0.8398, 1.1797, 0.9336, 1.0156, 0.9844, 1.0469, 0.9844, 0.9297, 1.1562, 1.1406, 1.4922, 1.125, 1.1641, 1.9453, 1.2344, 1.8828, 0.9414, 1.0859, 0.8164, 1.0078, 1.1875, 1.5547, 0.8164, 1.4453, 0.582, 1.3672, 1.4219, 1.1172, 1.4453, 0.8164, 1.0625, 0.8086, 1.2344, 1.3047, 1.5312, 1.1016, 1.3594, 1.5547, 1.1016, 1.125, 1.6094, 1.125, 0.7305, 1.1641, 0.9023, 1.1484, 0.8828, 0.9414, 1.0234, 1.0469, 1.1797, 1.0547, 0.7812, 1.7109, 1.0312, 1.0234, 1.3984, 0.8867, 1.0625, 1.3984, 1.0391, 0.8203, 1.1172, 1.0625, 1.0, 1.2188, 1.25, 1.0703, 1.1094, 1.1719, 0.9766, 1.0625, 1.3516, 0.8359, 1.3203, 1.0391, 1.4219, 1.5078, 1.4922, 1.4844, 0.7031, 1.4297, 1.0547, 1.0859, 0.9141, 0.6914, 1.125, 1.2109, 0.8867, 1.1172, 1.5391, 1.5625, 1.7188, 0.9258, 1.1406, 1.4453, 1.4609, 1.6172, 1.5625, 1.3984, 1.375, 1.4922, 1.3281, 1.8438, 0.9805, 1.0625, 1.5703, 0.9492, 0.7422, 2.4531, 1.375, 1.125, 1.0312, 1.0547, 1.1172, 1.7109, 2.1094, 1.0703, 1.2969, 1.1172, 1.5938, 1.5859, 1.3828, 0.9453, 1.2656, 1.0156, 1.8203, 0.9258, 1.0312, 1.4688, 0.9102, 0.9883, 1.4375, 0.8672, 1.1484, 1.4609, 1.5625, 0.6875, 0.8633, 0.75, 1.4062, 0.9023, 1.3906, 1.1328, 1.125, 1.2656, 0.9805, 1.0234, 1.5234, 1.3359, 1.1016, 1.0234, 0.707, 1.2422, 0.9727, 1.4531, 1.3594, 0.7461, 1.4922, 1.0234, 1.2812, 1.7656, 0.9336, 1.4844, 0.8516, 1.1016, 1.5234, 0.9531, 0.8164, 0.8555, 1.4219, 0.7773, 1.0625, 1.3281, 1.4297, 1.0156, 1.4062, 1.1797, 0.9141, 1.2422, 0.9727, 0.8867, 0.9961, 0.9062, 0.9805, 0.9961, 1.8516, 1.1016, 0.9102, 0.6406, 0.7891, 1.0547, 0.7422, 1.1406, 1.1484, 1.3281, 1.2109, 0.9258, 1.1172, 0.8828, 0.8594, 0.9688, 0.8086, 1.2656, 1.3203, 1.2891, 0.9492, 0.7461, 1.0234, 0.9648, 1.5312, 1.0703, 1.9844, 1.1562, 1.0156, 1.2578, 1.5312, 1.1016, 1.4062, 1.0234, 0.9609, 1.0625, 1.2812, 1.4141, 1.0859, 1.25, 1.2109, 1.375, 1.3125, 0.8828, 0.8672, 1.1953, 1.0469, 1.2188, 0.5781, 1.9453, 1.0859, 1.2266, 1.0938, 0.7031, 1.5391, 0.8359, 0.9023, 0.7812, 1.5312, 1.4609, 0.9531, 0.6602, 1.4766, 1.3516, 1.0703, 0.9492, 0.7031, 1.8516, 1.4141, 1.0156, 1.0859, 1.1406, 0.9531, 1.3672, 0.75, 1.1172, 1.5, 1.6875, 0.8594, 1.0703, 1.0938, 1.1484, 1.4609, 1.0859, 1.2656, 0.7812, 1.2109, 1.5078, 1.0938, 0.9844, 1.3516, 1.2578, 1.0, 0.6094, 1.0469, 0.9961, 1.0938, 0.8398, 1.1484, 0.8984, 1.0469, 1.4766, 1.4141, 1.5547, 0.918, 0.6289, 1.2266, 1.0, 1.375, 1.2188, 1.6641, 0.7969, 1.1172, 1.0312, 0.7539, 0.6875, 1.6484, 1.2969, 1.0547, 1.8828, 0.7773, 1.8281, 1.5938, 1.1953, 1.5234, 1.375, 1.2578, 1.2266, 1.375, 1.0781, 1.5, 1.0781, 1.0156, 0.9922, 1.3984, 1.6562, 0.8789, 2.1875, 0.7812, 0.9961, 1.5078, 0.8633, 0.8477, 1.0156, 0.75, 1.2422, 1.1328, 0.7266, 1.1172, 0.7812, 0.7344, 0.8516, 1.1094, 1.1484, 1.5078, 0.7891, 1.2812, 1.4453, 1.2891, 0.9922, 1.6797, 1.4141, 1.0, 0.6016, 1.3906, 0.7617, 1.2188, 2.0156, 0.9297, 1.2422, 1.375, 0.9609, 1.2109, 1.2031, 1.0469, 0.9297, 0.8633, 0.9023, 0.8984, 0.8711, 1.1562, 1.2578, 1.2891, 1.4688, 1.6484, 1.2188, 1.3125, 0.9648, 1.1094, 1.5391, 1.0703, 1.1562, 1.0703, 0.9062, 1.1094, 1.4531, 1.3281, 1.2344, 0.6445, 0.9141, 1.4609, 1.6641, 1.1406, 0.8516, 1.8516, 0.7695]
|