Model save
Browse files- README.md +107 -0
- model.safetensors +1 -1
README.md
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
base_model: nguyenvulebinh/wav2vec2-base-vietnamese-250h
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- wer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: w2v2_ablation_200epoch-with_ling_head-0drop-0load_best-best_on_tp0.025_tl10_fp0.001_fl16
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# w2v2_ablation_200epoch-with_ling_head-0drop-0load_best-best_on_tp0.025_tl10_fp0.001_fl16
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [nguyenvulebinh/wav2vec2-base-vietnamese-250h](https://huggingface.co/nguyenvulebinh/wav2vec2-base-vietnamese-250h) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.5364
|
| 21 |
+
- Wer: 0.1808
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 2e-05
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 32
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 4
|
| 46 |
+
- total_train_batch_size: 32
|
| 47 |
+
- total_eval_batch_size: 128
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 51 |
+
- num_epochs: 200
|
| 52 |
+
- mixed_precision_training: Native AMP
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 57 |
+
|:-------------:|:------:|:-----:|:---------------:|:------:|
|
| 58 |
+
| 5.0922 | 4.72 | 500 | 5.2296 | 1.0 |
|
| 59 |
+
| 4.3263 | 9.43 | 1000 | 5.4260 | 1.0 |
|
| 60 |
+
| 1.9702 | 14.15 | 1500 | 1.6761 | 0.4369 |
|
| 61 |
+
| 0.7013 | 18.87 | 2000 | 0.6799 | 0.2360 |
|
| 62 |
+
| 0.4391 | 23.58 | 2500 | 0.5237 | 0.1964 |
|
| 63 |
+
| 0.3015 | 28.3 | 3000 | 0.4437 | 0.1849 |
|
| 64 |
+
| 0.2416 | 33.02 | 3500 | 0.4311 | 0.2081 |
|
| 65 |
+
| 0.2057 | 37.74 | 4000 | 0.4202 | 0.1697 |
|
| 66 |
+
| 0.1714 | 42.45 | 4500 | 0.4270 | 0.1738 |
|
| 67 |
+
| 0.1812 | 47.17 | 5000 | 0.4467 | 0.1600 |
|
| 68 |
+
| 0.1498 | 51.89 | 5500 | 0.4322 | 0.2197 |
|
| 69 |
+
| 0.1255 | 56.6 | 6000 | 0.4408 | 0.1696 |
|
| 70 |
+
| 0.1148 | 61.32 | 6500 | 0.4531 | 0.1765 |
|
| 71 |
+
| 0.1112 | 66.04 | 7000 | 0.4572 | 0.2148 |
|
| 72 |
+
| 0.1038 | 70.75 | 7500 | 0.4648 | 0.1894 |
|
| 73 |
+
| 0.0923 | 75.47 | 8000 | 0.4812 | 0.1558 |
|
| 74 |
+
| 0.086 | 80.19 | 8500 | 0.4882 | 0.1894 |
|
| 75 |
+
| 0.0872 | 84.91 | 9000 | 0.4662 | 0.1744 |
|
| 76 |
+
| 0.0778 | 89.62 | 9500 | 0.4800 | 0.1750 |
|
| 77 |
+
| 0.0709 | 94.34 | 10000 | 0.5077 | 0.1960 |
|
| 78 |
+
| 0.0703 | 99.06 | 10500 | 0.5038 | 0.1740 |
|
| 79 |
+
| 0.0721 | 103.77 | 11000 | 0.5131 | 0.1763 |
|
| 80 |
+
| 0.0717 | 108.49 | 11500 | 0.5091 | 0.1896 |
|
| 81 |
+
| 0.0818 | 113.21 | 12000 | 0.5173 | 0.1908 |
|
| 82 |
+
| 0.0626 | 117.92 | 12500 | 0.5158 | 0.1865 |
|
| 83 |
+
| 0.0749 | 122.64 | 13000 | 0.5208 | 0.1865 |
|
| 84 |
+
| 0.0592 | 127.36 | 13500 | 0.5244 | 0.1781 |
|
| 85 |
+
| 0.055 | 132.08 | 14000 | 0.5303 | 0.1810 |
|
| 86 |
+
| 0.0487 | 136.79 | 14500 | 0.5264 | 0.1739 |
|
| 87 |
+
| 0.0486 | 141.51 | 15000 | 0.5225 | 0.1814 |
|
| 88 |
+
| 0.0478 | 146.23 | 15500 | 0.5316 | 0.1870 |
|
| 89 |
+
| 0.0453 | 150.94 | 16000 | 0.5270 | 0.1776 |
|
| 90 |
+
| 0.0449 | 155.66 | 16500 | 0.5318 | 0.1821 |
|
| 91 |
+
| 0.0585 | 160.38 | 17000 | 0.5332 | 0.1775 |
|
| 92 |
+
| 0.0481 | 165.09 | 17500 | 0.5373 | 0.1784 |
|
| 93 |
+
| 0.0459 | 169.81 | 18000 | 0.5335 | 0.1756 |
|
| 94 |
+
| 0.0473 | 174.53 | 18500 | 0.5360 | 0.1808 |
|
| 95 |
+
| 0.0512 | 179.25 | 19000 | 0.5347 | 0.1791 |
|
| 96 |
+
| 0.046 | 183.96 | 19500 | 0.5367 | 0.1778 |
|
| 97 |
+
| 0.048 | 188.68 | 20000 | 0.5354 | 0.1783 |
|
| 98 |
+
| 0.0471 | 193.4 | 20500 | 0.5366 | 0.1814 |
|
| 99 |
+
| 0.0419 | 198.11 | 21000 | 0.5364 | 0.1808 |
|
| 100 |
+
|
| 101 |
+
|
| 102 |
+
### Framework versions
|
| 103 |
+
|
| 104 |
+
- Transformers 4.35.2
|
| 105 |
+
- Pytorch 1.13.1+cu117
|
| 106 |
+
- Datasets 2.12.0
|
| 107 |
+
- Tokenizers 0.14.1
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 197617854
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5652ae2d91ad6f8b5850c9ab79ead527d59294d16080b09266358d925097136f
|
| 3 |
size 197617854
|