update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
metrics:
|
| 5 |
+
- accuracy
|
| 6 |
+
model-index:
|
| 7 |
+
- name: TED_CLM_gpt2_tedlium_additional_head
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
# TED_CLM_gpt2_tedlium_additional_head
|
| 15 |
+
|
| 16 |
+
This model is a fine-tuned version of [Lakoc/gpt2_512h_16l_add_head8](https://huggingface.co/Lakoc/gpt2_512h_16l_add_head8) on the None dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 1.9139
|
| 19 |
+
- Accuracy: 0.5529
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 0.001
|
| 39 |
+
- train_batch_size: 128
|
| 40 |
+
- eval_batch_size: 128
|
| 41 |
+
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 4
|
| 44 |
+
- total_train_batch_size: 512
|
| 45 |
+
- total_eval_batch_size: 512
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: linear
|
| 48 |
+
- lr_scheduler_warmup_steps: 20000
|
| 49 |
+
- num_epochs: 15.0
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 54 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 55 |
+
| 2.2945 | 0.62 | 3000 | 2.4760 | 0.4352 |
|
| 56 |
+
| 2.0669 | 1.24 | 6000 | 2.2729 | 0.4767 |
|
| 57 |
+
| 1.9754 | 1.86 | 9000 | 2.1827 | 0.4974 |
|
| 58 |
+
| 1.9292 | 2.49 | 12000 | 2.1086 | 0.5139 |
|
| 59 |
+
| 1.8983 | 3.11 | 15000 | 2.0666 | 0.5223 |
|
| 60 |
+
| 1.8853 | 3.73 | 18000 | 2.0389 | 0.5278 |
|
| 61 |
+
| 1.8708 | 4.35 | 21000 | 2.0216 | 0.5301 |
|
| 62 |
+
| 1.8524 | 4.97 | 24000 | 2.0024 | 0.5352 |
|
| 63 |
+
| 1.836 | 5.59 | 27000 | 1.9915 | 0.5365 |
|
| 64 |
+
| 1.8219 | 6.22 | 30000 | 1.9847 | 0.5410 |
|
| 65 |
+
| 1.8134 | 6.84 | 33000 | 1.9670 | 0.5408 |
|
| 66 |
+
| 1.8088 | 7.46 | 36000 | 1.9736 | 0.5425 |
|
| 67 |
+
| 1.8011 | 8.08 | 39000 | 1.9610 | 0.5426 |
|
| 68 |
+
| 1.7901 | 8.7 | 42000 | 1.9519 | 0.5459 |
|
| 69 |
+
| 1.7829 | 9.32 | 45000 | 1.9524 | 0.5463 |
|
| 70 |
+
| 1.7865 | 9.94 | 48000 | 1.9424 | 0.5479 |
|
| 71 |
+
| 1.7775 | 10.57 | 51000 | 1.9421 | 0.5480 |
|
| 72 |
+
| 1.7698 | 11.19 | 54000 | 1.9346 | 0.5486 |
|
| 73 |
+
| 1.767 | 11.81 | 57000 | 1.9249 | 0.5493 |
|
| 74 |
+
| 1.7578 | 12.43 | 60000 | 1.9262 | 0.5500 |
|
| 75 |
+
| 1.7613 | 13.05 | 63000 | 1.9185 | 0.5508 |
|
| 76 |
+
| 1.7591 | 13.67 | 66000 | 1.9191 | 0.5523 |
|
| 77 |
+
| 1.7489 | 14.29 | 69000 | 1.9159 | 0.5522 |
|
| 78 |
+
| 1.7506 | 14.92 | 72000 | 1.9139 | 0.5529 |
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
### Framework versions
|
| 82 |
+
|
| 83 |
+
- Transformers 4.31.0.dev0
|
| 84 |
+
- Pytorch 2.1.0+cu121
|
| 85 |
+
- Datasets 2.13.1
|
| 86 |
+
- Tokenizers 0.13.3
|