Commit
·
cf937a0
1
Parent(s):
de8ef08
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- timit_asr
|
| 7 |
+
model-index:
|
| 8 |
+
- name: test_last_transformer_1
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# test_last_transformer_1
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the timit_asr dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.3254
|
| 20 |
+
- Cer: 0.1178
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 0.0001
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- lr_scheduler_warmup_steps: 1000
|
| 46 |
+
- num_epochs: 10
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Cer |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 53 |
+
| 6.5814 | 0.4 | 100 | 3.4669 | 0.8825 |
|
| 54 |
+
| 2.9541 | 0.8 | 200 | 2.1664 | 0.6831 |
|
| 55 |
+
| 1.313 | 1.2 | 300 | 0.7007 | 0.2299 |
|
| 56 |
+
| 0.6578 | 1.61 | 400 | 0.4822 | 0.1815 |
|
| 57 |
+
| 0.5294 | 2.01 | 500 | 0.4487 | 0.1603 |
|
| 58 |
+
| 0.4607 | 2.41 | 600 | 0.3862 | 0.1441 |
|
| 59 |
+
| 0.4404 | 2.81 | 700 | 0.3722 | 0.1443 |
|
| 60 |
+
| 0.4014 | 3.21 | 800 | 0.3643 | 0.1335 |
|
| 61 |
+
| 0.4038 | 3.61 | 900 | 0.3462 | 0.1306 |
|
| 62 |
+
| 0.3876 | 4.02 | 1000 | 0.3295 | 0.1304 |
|
| 63 |
+
| 0.3461 | 4.42 | 1100 | 0.3139 | 0.1254 |
|
| 64 |
+
| 0.3414 | 4.82 | 1200 | 0.3010 | 0.1231 |
|
| 65 |
+
| 0.3096 | 5.22 | 1300 | 0.3078 | 0.1230 |
|
| 66 |
+
| 0.305 | 5.62 | 1400 | 0.3273 | 0.1299 |
|
| 67 |
+
| 0.2868 | 6.02 | 1500 | 0.3016 | 0.1214 |
|
| 68 |
+
| 0.2535 | 6.43 | 1600 | 0.3022 | 0.1194 |
|
| 69 |
+
| 0.2596 | 6.83 | 1700 | 0.2980 | 0.1209 |
|
| 70 |
+
| 0.2414 | 7.23 | 1800 | 0.3130 | 0.1200 |
|
| 71 |
+
| 0.2174 | 7.63 | 1900 | 0.3076 | 0.1178 |
|
| 72 |
+
| 0.2214 | 8.03 | 2000 | 0.3021 | 0.1174 |
|
| 73 |
+
| 0.197 | 8.43 | 2100 | 0.3110 | 0.1182 |
|
| 74 |
+
| 0.193 | 8.84 | 2200 | 0.3169 | 0.1182 |
|
| 75 |
+
| 0.187 | 9.24 | 2300 | 0.3187 | 0.1185 |
|
| 76 |
+
| 0.1772 | 9.64 | 2400 | 0.3254 | 0.1178 |
|
| 77 |
+
|
| 78 |
+
|
| 79 |
+
### Framework versions
|
| 80 |
+
|
| 81 |
+
- Transformers 4.17.0
|
| 82 |
+
- Pytorch 2.4.0
|
| 83 |
+
- Datasets 1.18.3
|
| 84 |
+
- Tokenizers 0.20.3
|