Hugging Face's logo

amd
/
SSD-1B-amdnpu

Text-to-Image
ONNX
Safetensors
English
stable-diffusion
stable-diffusion-diffusers
distilled-model
RyzenAI
Quantization
ONNX
Computer Vision
Model card Files
xet
 Community
huisu commited on
Commit
961a58e
·
verified ·
1 Parent(s): e742cb9

Upload 27 files

Browse files
Files changed (28) hide show
  1. .gitattributes +4 -0
  2. scheduler/scheduler_config.json +18 -0
  3. text_encoder/config.json +24 -0
  4. text_encoder/model.safetensors +3 -0
  5. text_encoder_2/config.json +24 -0
  6. text_encoder_2/model.safetensors +3 -0
  7. tokenizer/merges.txt +0 -0
  8. tokenizer/special_tokens_map.json +24 -0
  9. tokenizer/tokenizer_config.json +33 -0
  10. tokenizer/vocab.json +0 -0
  11. tokenizer_2/merges.txt +0 -0
  12. tokenizer_2/special_tokens_map.json +24 -0
  13. tokenizer_2/tokenizer_config.json +33 -0
  14. tokenizer_2/vocab.json +0 -0
  15. unet/config.json +74 -0
  16. unet/dd/cache/NhwcConv_0-unetconv_inConv_meta.json +0 -0
  17. unet/dd/dd_metastate_SD15_Unet_NhwcConv_0-unetconv_inConv.ddbundle +3 -0
  18. unet/dd/dd_metastate_SD15_Unet_NhwcConv_0-unetconv_inConv.fconst +3 -0
  19. unet/dd/dd_metastate_SD15_Unet_NhwcConv_0-unetconv_inConv.state +0 -0
  20. unet/dd/onnx_report.txt +104 -0
  21. unet/dd/replaced.onnx +3 -0
  22. vae_decoder/config.json +31 -0
  23. vae_decoder/dd/cache/NhwcConv_0-post_quant_convConv_meta.json +0 -0
  24. vae_decoder/dd/dd_metastate_Sd15_Decoder_NhwcConv_0-post_quant_convConv.ddbundle +3 -0
  25. vae_decoder/dd/dd_metastate_Sd15_Decoder_NhwcConv_0-post_quant_convConv.fconst +3 -0
  26. vae_decoder/dd/dd_metastate_Sd15_Decoder_NhwcConv_0-post_quant_convConv.state +0 -0
  27. vae_decoder/dd/onnx_report.txt +36 -0
  28. vae_decoder/dd/replaced.onnx +3 -0
.gitattributes CHANGED
@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ unet/dd/dd_metastate_SD15_Unet_NhwcConv_0-unetconv_inConv.ddbundle filter=lfs diff=lfs merge=lfs -text
37
+ unet/dd/dd_metastate_SD15_Unet_NhwcConv_0-unetconv_inConv.fconst filter=lfs diff=lfs merge=lfs -text
38
+ vae_decoder/dd/dd_metastate_Sd15_Decoder_NhwcConv_0-post_quant_convConv.ddbundle filter=lfs diff=lfs merge=lfs -text
39
+ vae_decoder/dd/dd_metastate_Sd15_Decoder_NhwcConv_0-post_quant_convConv.fconst filter=lfs diff=lfs merge=lfs -text
scheduler/scheduler_config.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "EulerDiscreteScheduler",
3
+ "_diffusers_version": "0.19.0",
4
+ "beta_end": 0.012,
5
+ "beta_schedule": "scaled_linear",
6
+ "beta_start": 0.00085,
7
+ "clip_sample": false,
8
+ "interpolation_type": "linear",
9
+ "num_train_timesteps": 1000,
10
+ "prediction_type": "epsilon",
11
+ "sample_max_value": 1.0,
12
+ "set_alpha_to_one": false,
13
+ "skip_prk_steps": true,
14
+ "steps_offset": 1,
15
+ "timestep_spacing": "leading",
16
+ "trained_betas": null,
17
+ "use_karras_sigmas": false
18
+ }
text_encoder/config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPTextModel"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 0,
7
+ "dropout": 0.0,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "quick_gelu",
10
+ "hidden_size": 768,
11
+ "initializer_factor": 1.0,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-05,
15
+ "max_position_embeddings": 77,
16
+ "model_type": "clip_text_model",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 1,
20
+ "projection_dim": 768,
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.29.2",
23
+ "vocab_size": 49408
24
+ }
text_encoder/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8377b1ca9d88fe06ec483dd7b3cfc62e5e8dbf8ddd252f455e79d659fa0553c5
3
+ size 492265880
text_encoder_2/config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "CLIPTextModelWithProjection"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 0,
7
+ "dropout": 0.0,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "gelu",
10
+ "hidden_size": 1280,
11
+ "initializer_factor": 1.0,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 5120,
14
+ "layer_norm_eps": 1e-05,
15
+ "max_position_embeddings": 77,
16
+ "model_type": "clip_text_model",
17
+ "num_attention_heads": 20,
18
+ "num_hidden_layers": 32,
19
+ "pad_token_id": 1,
20
+ "projection_dim": 1280,
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.29.2",
23
+ "vocab_size": 49408
24
+ }
text_encoder_2/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b84f413eebecbd049b72874c1df533a516510cb5a2489ae58c7e320209cf0ebe
3
+ size 2778702976
tokenizer/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|startoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<|endoftext|>",
17
+ "unk_token": {
18
+ "content": "<|endoftext|>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": {
4
+ "__type": "AddedToken",
5
+ "content": "<|startoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "clean_up_tokenization_spaces": true,
12
+ "do_lower_case": true,
13
+ "eos_token": {
14
+ "__type": "AddedToken",
15
+ "content": "<|endoftext|>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "errors": "replace",
22
+ "model_max_length": 77,
23
+ "pad_token": "<|endoftext|>",
24
+ "tokenizer_class": "CLIPTokenizer",
25
+ "unk_token": {
26
+ "__type": "AddedToken",
27
+ "content": "<|endoftext|>",
28
+ "lstrip": false,
29
+ "normalized": true,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
tokenizer/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_2/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_2/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|startoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "!",
17
+ "unk_token": {
18
+ "content": "<|endoftext|>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer_2/tokenizer_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": {
4
+ "__type": "AddedToken",
5
+ "content": "<|startoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "clean_up_tokenization_spaces": true,
12
+ "do_lower_case": true,
13
+ "eos_token": {
14
+ "__type": "AddedToken",
15
+ "content": "<|endoftext|>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "errors": "replace",
22
+ "model_max_length": 77,
23
+ "pad_token": "!",
24
+ "tokenizer_class": "CLIPTokenizer",
25
+ "unk_token": {
26
+ "__type": "AddedToken",
27
+ "content": "<|endoftext|>",
28
+ "lstrip": false,
29
+ "normalized": true,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
tokenizer_2/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
unet/config.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.19.0.dev0",
4
+ "act_fn": "silu",
5
+ "addition_embed_type": "text_time",
6
+ "addition_embed_type_num_heads": 64,
7
+ "addition_time_embed_dim": 256,
8
+ "attention_head_dim": [
9
+ 5,
10
+ 10,
11
+ 20
12
+ ],
13
+ "block_out_channels": [
14
+ 320,
15
+ 640,
16
+ 1280
17
+ ],
18
+ "center_input_sample": false,
19
+ "class_embed_type": null,
20
+ "class_embeddings_concat": false,
21
+ "conv_in_kernel": 3,
22
+ "conv_out_kernel": 3,
23
+ "cross_attention_dim": 2048,
24
+ "cross_attention_norm": null,
25
+ "down_block_types": [
26
+ "DownBlock2D",
27
+ "CrossAttnDownBlock2D",
28
+ "CrossAttnDownBlock2D"
29
+ ],
30
+ "downsample_padding": 1,
31
+ "dual_cross_attention": false,
32
+ "encoder_hid_dim": null,
33
+ "encoder_hid_dim_type": null,
34
+ "flip_sin_to_cos": true,
35
+ "freq_shift": 0,
36
+ "in_channels": 4,
37
+ "layers_per_block": 2,
38
+ "mid_block_only_cross_attention": null,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2D",
41
+ "norm_eps": 1e-05,
42
+ "norm_num_groups": 32,
43
+ "num_attention_heads": null,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "out_channels": 4,
47
+ "projection_class_embeddings_input_dim": 2816,
48
+ "resnet_out_scale_factor": 1.0,
49
+ "resnet_skip_time_act": false,
50
+ "resnet_time_scale_shift": "default",
51
+ "sample_size": 128,
52
+ "time_cond_proj_dim": null,
53
+ "time_embedding_act_fn": null,
54
+ "time_embedding_dim": null,
55
+ "time_embedding_type": "positional",
56
+ "timestep_post_act": null,
57
+ "transformer_layers_per_block": [
58
+ [1],
59
+ [2,2],
60
+ [4,4]
61
+ ],
62
+ "reverse_transformer_layers_per_block": [
63
+ [4,4,10],
64
+ [2,1,1],
65
+ 1
66
+ ],
67
+ "up_block_types": [
68
+ "CrossAttnUpBlock2D",
69
+ "CrossAttnUpBlock2D",
70
+ "UpBlock2D"
71
+ ],
72
+ "upcast_attention": null,
73
+ "use_linear_projection": true
74
+ }
unet/dd/cache/NhwcConv_0-unetconv_inConv_meta.json ADDED
The diff for this file is too large to render. See raw diff
 
unet/dd/dd_metastate_SD15_Unet_NhwcConv_0-unetconv_inConv.ddbundle ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac1138a7d6ea40499328e07d093e481776346dd35c7fc7324e4838a2d2dcb0ce
3
+ size 1394734577
unet/dd/dd_metastate_SD15_Unet_NhwcConv_0-unetconv_inConv.fconst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f56a8c65e8ff326f0dcb43df953ba6466aa29ebcbbf003c56a20224106cfee6
3
+ size 1547034688
unet/dd/dd_metastate_SD15_Unet_NhwcConv_0-unetconv_inConv.state ADDED
Binary file (7.06 kB). View file
 
unet/dd/onnx_report.txt ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ DynamicDispatch Offload - not offloaded
2
+ +-------------------+-------+------------------------+------------------------+
3
+ | Op Type | Count | Inputs | Outputs |
4
+ +===================+=======+========================+========================+
5
+ | Cast | 1 | ['batch',1] - INT64 | ['batch',1] - FLOAT |
6
+ | CastAvx | 1 | ['batch','height','wi� | ['batch','height','wi� |
7
+ | | | - FLOAT | - BFLOAT16 |
8
+ | CastAvx | 1 | ['batch',77,2048] - | ['batch',77,2048] - |
9
+ | | | FLOAT | BFLOAT16 |
10
+ | CastAvx | 1 | ['batch',2816] - FLOAT | ['batch',2816] - |
11
+ | | | | BFLOAT16 |
12
+ | CastAvx | 1 | ['batch',320] - FLOAT | ['batch',320] - |
13
+ | | | | BFLOAT16 |
14
+ | CastAvx | 1 | ['batch','height','wi� | ['batch','height','wi� |
15
+ | | | - BFLOAT16 | - FLOAT |
16
+ | Concat | 1 | [1] - INT64 | [1] - INT64 |
17
+ | Concat | 1 | [1] - INT64 | [2] - INT64 |
18
+ | | | [1] - INT64 | |
19
+ | Concat | 1 | ['6*batch',128] - | ['6*batch',256] - |
20
+ | | | FLOAT | FLOAT |
21
+ | | | ['6*batch',128] - | |
22
+ | | | FLOAT | |
23
+ | Concat | 1 | ['batch',1280] - FLOAT | ['batch',2816] - FLOAT |
24
+ | | | ['batch',1536] - FLOAT | |
25
+ | Concat | 1 | ['batch',160] - FLOAT | ['batch',320] - FLOAT |
26
+ | | | ['batch',160] - FLOAT | |
27
+ | Cos | 1 | ['6*batch',128] - | ['6*batch',128] - |
28
+ | | | FLOAT | FLOAT |
29
+ | Cos | 1 | ['batch',160] - FLOAT | ['batch',160] - FLOAT |
30
+ | Equal | 1 | [1] - INT64 | [1] - BOOL |
31
+ | | | [1] - INT64 | |
32
+ | Expand | 1 | [1] - INT64 | ['batch'] - INT64 |
33
+ | | | [1] - INT64 | |
34
+ | Gather | 1 | [4] - INT64 | [1] - INT64 |
35
+ | | | [1] - INT64 | |
36
+ | Gather | 1 | [2] - INT64 | [1] - INT64 |
37
+ | | | [1] - INT64 | |
38
+ | Mul | 1 | ['6*batch',1] - FLOAT | ['6*batch',128] - |
39
+ | | | [1,128] - FLOAT | FLOAT |
40
+ | Mul | 1 | ['batch',1] - FLOAT | ['batch',160] - FLOAT |
41
+ | | | [1,160] - FLOAT | |
42
+ | Reshape | 1 | ['batch',6] - FLOAT | ['6*batch',1] - FLOAT |
43
+ | | | [2] - INT64 | |
44
+ | Reshape | 2 | [1] - INT64 | [1] - INT64 |
45
+ | | | [1] - INT64 | |
46
+ | Reshape | 1 | ['batch'] - INT64 | ['batch',1] - INT64 |
47
+ | | | [2] - INT64 | |
48
+ | Reshape | 1 | ['6*batch',256] - | ['batch',1536] - FLOAT |
49
+ | | | FLOAT | |
50
+ | | | [2] - INT64 | |
51
+ | Shape | 1 | ['batch',4,'height','� | [4] - INT64 |
52
+ | | | - FLOAT | |
53
+ | Shape | 1 | ['batch',1280] - FLOAT | [2] - INT64 |
54
+ | Sin | 1 | ['6*batch',128] - | ['6*batch',128] - |
55
+ | | | FLOAT | FLOAT |
56
+ | Sin | 1 | ['batch',160] - FLOAT | ['batch',160] - FLOAT |
57
+ | Transpose | 1 | ['batch',4,'height','� | ['batch','height','wi� |
58
+ | | | - FLOAT | - FLOAT |
59
+ | Transpose | 1 | ['batch','height','wi� | ['batch',4,'height','� |
60
+ | | | - FLOAT | - FLOAT |
61
+ | Where | 1 | [1] - BOOL | [1] - INT64 |
62
+ | | | [1] - INT64 | |
63
+ | | | [1] - INT64 | |
64
+ +-------------------+-------+------------------------+------------------------+
65
+ | Not offloaded sum | 31 | | |
66
+ +-------------------+-------+------------------------+------------------------+
67
+ DynamicDispatch Offload - offloaded
68
+ +-------------------+-----------------+-------------------+-------------------+
69
+ | Op Type | Count | Inputs | Outputs |
70
+ +===================+=================+===================+===================+
71
+ | DynamicDispatch | 1 | ['batch','height� | ['batch','height� |
72
+ | | | - BFLOAT16 | - BFLOAT16 |
73
+ | | | ['batch',2816] - | |
74
+ | | | BFLOAT16 | |
75
+ | | | ['batch',77,2048] | |
76
+ | | | - BFLOAT16 | |
77
+ | | | ['batch',320] - | |
78
+ | | | BFLOAT16 | |
79
+ +-------------------+-----------------+-------------------+-------------------+
80
+ | Offloaded sum | 1 | | |
81
+ +-------------------+-----------------+-------------------+-------------------+
82
+ | Offloaded Op | SDAdd | | |
83
+ | Types | SDMul | | |
84
+ | | SDConcat | | |
85
+ | | SDFlatMHA | | |
86
+ | | SDMHA_bfp | | |
87
+ | | SDSilu | | |
88
+ | | SDGemm | | |
89
+ | | SDLayerNorm_bfp | | |
90
+ | | SDResize | | |
91
+ | | SDMul_bfp | | |
92
+ | | SDGroupNorm | | |
93
+ | | SDCastBfp2Bf | | |
94
+ | | SDCastBf2Bfp | | |
95
+ | | SDAdd_bfp | | |
96
+ | | SDLayerNorm | | |
97
+ | | SDGemm_bfp | | |
98
+ | | SDConv | | |
99
+ +-------------------+-----------------+-------------------+-------------------+
100
+ | Offloaded sum (dd | 895 | | |
101
+ | fusion) | | | |
102
+ | Offload Ratio (dd | 96.65% | | |
103
+ | fusion) | | | |
104
+ +-------------------+-----------------+-------------------+-------------------+
unet/dd/replaced.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3599cda6c83ab06cb4f5263a06cb68eb0ab6f6aefa673f5340bafddb0662143b
3
+ size 816249
vae_decoder/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AutoencoderKL",
3
+ "_diffusers_version": "0.19.0",
4
+ "act_fn": "silu",
5
+ "block_out_channels": [
6
+ 128,
7
+ 256,
8
+ 512,
9
+ 512
10
+ ],
11
+ "down_block_types": [
12
+ "DownEncoderBlock2D",
13
+ "DownEncoderBlock2D",
14
+ "DownEncoderBlock2D",
15
+ "DownEncoderBlock2D"
16
+ ],
17
+ "force_upcast": true,
18
+ "in_channels": 3,
19
+ "latent_channels": 4,
20
+ "layers_per_block": 2,
21
+ "norm_num_groups": 32,
22
+ "out_channels": 3,
23
+ "sample_size": 1024,
24
+ "scaling_factor": 0.13025,
25
+ "up_block_types": [
26
+ "UpDecoderBlock2D",
27
+ "UpDecoderBlock2D",
28
+ "UpDecoderBlock2D",
29
+ "UpDecoderBlock2D"
30
+ ]
31
+ }
vae_decoder/dd/cache/NhwcConv_0-post_quant_convConv_meta.json ADDED
The diff for this file is too large to render. See raw diff
 
vae_decoder/dd/dd_metastate_Sd15_Decoder_NhwcConv_0-post_quant_convConv.ddbundle ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f00f4af008a3523298f39d0f14283a343a09ed5e694fb7acf2016a6cb45e5530
3
+ size 717595506
vae_decoder/dd/dd_metastate_Sd15_Decoder_NhwcConv_0-post_quant_convConv.fconst ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d11b05d368ecf8c419feeb7a959b8082f572b58c066157dda5a8df379781a1b
3
+ size 61113872
vae_decoder/dd/dd_metastate_Sd15_Decoder_NhwcConv_0-post_quant_convConv.state ADDED
Binary file (5.41 kB). View file
 
vae_decoder/dd/onnx_report.txt ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ DynamicDispatch Offload - not offloaded
2
+ +-------------------+-------+------------------------+------------------------+
3
+ | Op Type | Count | Inputs | Outputs |
4
+ +===================+=======+========================+========================+
5
+ | CastAvx | 1 | [1,'height','width',4] | [1,'height','width',4] |
6
+ | | | - FLOAT | - BFLOAT16 |
7
+ | CastAvx | 1 | [1,'floor(2.0*floor(2� | [1,'floor(2.0*floor(2� |
8
+ | | | - BFLOAT16 | - FLOAT |
9
+ | Transpose | 1 | [1,4,'height','width'] | [1,'height','width',4] |
10
+ | | | - FLOAT | - FLOAT |
11
+ | Transpose | 1 | [1,'floor(2.0*floor(2� | [1,3,'floor(2.0*floor� |
12
+ | | | - FLOAT | - FLOAT |
13
+ +-------------------+-------+------------------------+------------------------+
14
+ | Not offloaded sum | 4 | | |
15
+ +-------------------+-------+------------------------+------------------------+
16
+ DynamicDispatch Offload - offloaded
17
+ +--------------------+-------------+---------------------+--------------------+
18
+ | Op Type | Count | Inputs | Outputs |
19
+ +====================+=============+=====================+====================+
20
+ | DynamicDispatch | 1 | [1,'height','width� | [1,'floor(2.0*flo� |
21
+ | | | - BFLOAT16 | - BFLOAT16 |
22
+ +--------------------+-------------+---------------------+--------------------+
23
+ | Offloaded sum | 1 | | |
24
+ +--------------------+-------------+---------------------+--------------------+
25
+ | Offloaded Op Types | SDGroupNorm | | |
26
+ | | SDResize | | |
27
+ | | SDMHA_VAE | | |
28
+ | | SDGemm | | |
29
+ | | SDConv | | |
30
+ | | SDAdd | | |
31
+ +--------------------+-------------+---------------------+--------------------+
32
+ | Offloaded sum (dd | 89 | | |
33
+ | fusion) | | | |
34
+ | Offload Ratio (dd | 95.7% | | |
35
+ | fusion) | | | |
36
+ +--------------------+-------------+---------------------+--------------------+
vae_decoder/dd/replaced.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:188300511a830dd0f8f5db137e5524fa6bd233050f402a1cf57f70bf32ecdbbb
3
+ size 47015