update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
metrics:
|
| 5 |
+
- wer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: hubert-base-libri-pruning-TEST15
|
| 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 |
+
# hubert-base-libri-pruning-TEST15
|
| 15 |
+
|
| 16 |
+
This model was trained from scratch on the None dataset.
|
| 17 |
+
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: -0.1647
|
| 19 |
+
- Wer: 0.1120
|
| 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: 0.00015
|
| 39 |
+
- train_batch_size: 64
|
| 40 |
+
- eval_batch_size: 8
|
| 41 |
+
- seed: 42
|
| 42 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
+
- lr_scheduler_type: linear
|
| 44 |
+
- lr_scheduler_warmup_steps: 3000
|
| 45 |
+
- num_epochs: 30
|
| 46 |
+
- mixed_precision_training: Native AMP
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 51 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 52 |
+
| 0.1439 | 1.12 | 500 | 0.1094 | 0.1124 |
|
| 53 |
+
| 0.1385 | 2.24 | 1000 | 0.1164 | 0.1121 |
|
| 54 |
+
| 0.1382 | 3.36 | 1500 | 0.1255 | 0.1124 |
|
| 55 |
+
| 0.1471 | 4.48 | 2000 | 0.1223 | 0.1117 |
|
| 56 |
+
| 0.1273 | 5.61 | 2500 | 0.0958 | 0.1121 |
|
| 57 |
+
| 0.0876 | 6.73 | 3000 | 0.0712 | 0.1120 |
|
| 58 |
+
| 0.067 | 7.85 | 3500 | 0.0461 | 0.1121 |
|
| 59 |
+
| 0.0502 | 8.97 | 4000 | 0.0251 | 0.1119 |
|
| 60 |
+
| 0.0279 | 10.09 | 4500 | 0.0051 | 0.1123 |
|
| 61 |
+
| -0.003 | 11.21 | 5000 | -0.0139 | 0.1123 |
|
| 62 |
+
| -0.016 | 12.33 | 5500 | -0.0303 | 0.1117 |
|
| 63 |
+
| -0.0375 | 13.45 | 6000 | -0.0479 | 0.1118 |
|
| 64 |
+
| -0.0515 | 14.57 | 6500 | -0.0630 | 0.1124 |
|
| 65 |
+
| -0.0578 | 15.7 | 7000 | -0.0768 | 0.1123 |
|
| 66 |
+
| -0.0727 | 16.82 | 7500 | -0.0911 | 0.1123 |
|
| 67 |
+
| -0.0854 | 17.94 | 8000 | -0.1032 | 0.1123 |
|
| 68 |
+
| -0.0987 | 19.06 | 8500 | -0.1132 | 0.1123 |
|
| 69 |
+
| -0.1018 | 20.18 | 9000 | -0.1225 | 0.1122 |
|
| 70 |
+
| -0.1129 | 21.3 | 9500 | -0.1321 | 0.1123 |
|
| 71 |
+
| -0.1252 | 22.42 | 10000 | -0.1399 | 0.1121 |
|
| 72 |
+
| -0.1237 | 23.54 | 10500 | -0.1468 | 0.1120 |
|
| 73 |
+
| -0.1316 | 24.66 | 11000 | -0.1523 | 0.1122 |
|
| 74 |
+
| -0.1317 | 25.78 | 11500 | -0.1571 | 0.1120 |
|
| 75 |
+
| -0.1445 | 26.91 | 12000 | -0.1610 | 0.1123 |
|
| 76 |
+
| -0.1393 | 28.03 | 12500 | -0.1635 | 0.1120 |
|
| 77 |
+
| -0.1453 | 29.15 | 13000 | -0.1647 | 0.1120 |
|
| 78 |
+
|
| 79 |
+
|
| 80 |
+
### Framework versions
|
| 81 |
+
|
| 82 |
+
- Transformers 4.30.0.dev0
|
| 83 |
+
- Pytorch 2.0.1
|
| 84 |
+
- Datasets 2.12.1.dev0
|
| 85 |
+
- Tokenizers 0.13.3
|