Commit
•
3791502
1
Parent(s):
3c00acb
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- common_voice
|
| 7 |
+
model-index:
|
| 8 |
+
- name: ''
|
| 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 |
+
#
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.0360
|
| 20 |
+
- Wer: 0.6551
|
| 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: 0.0004
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
+
- eval_batch_size: 32
|
| 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: 800
|
| 46 |
+
- num_epochs: 200.0
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
+
|:-------------:|:------:|:-----:|:---------------:|:------:|
|
| 53 |
+
| 5.649 | 9.62 | 500 | 3.0038 | 1.0 |
|
| 54 |
+
| 1.6272 | 19.23 | 1000 | 0.7362 | 0.7819 |
|
| 55 |
+
| 1.1354 | 28.85 | 1500 | 0.6410 | 0.7111 |
|
| 56 |
+
| 1.0424 | 38.46 | 2000 | 0.6907 | 0.7431 |
|
| 57 |
+
| 0.9293 | 48.08 | 2500 | 0.7249 | 0.7102 |
|
| 58 |
+
| 0.8246 | 57.69 | 3000 | 0.7422 | 0.6966 |
|
| 59 |
+
| 0.7837 | 67.31 | 3500 | 0.7413 | 0.6813 |
|
| 60 |
+
| 0.7147 | 76.92 | 4000 | 0.7873 | 0.6930 |
|
| 61 |
+
| 0.6276 | 86.54 | 4500 | 0.8038 | 0.6677 |
|
| 62 |
+
| 0.6041 | 96.15 | 5000 | 0.8240 | 0.6831 |
|
| 63 |
+
| 0.5336 | 105.77 | 5500 | 0.8748 | 0.6749 |
|
| 64 |
+
| 0.4705 | 115.38 | 6000 | 0.9006 | 0.6497 |
|
| 65 |
+
| 0.43 | 125.0 | 6500 | 0.8954 | 0.6551 |
|
| 66 |
+
| 0.3859 | 134.62 | 7000 | 0.9074 | 0.6614 |
|
| 67 |
+
| 0.3342 | 144.23 | 7500 | 0.9693 | 0.6560 |
|
| 68 |
+
| 0.3155 | 153.85 | 8000 | 1.0073 | 0.6691 |
|
| 69 |
+
| 0.2673 | 163.46 | 8500 | 1.0170 | 0.6632 |
|
| 70 |
+
| 0.2409 | 173.08 | 9000 | 1.0304 | 0.6709 |
|
| 71 |
+
| 0.2189 | 182.69 | 9500 | 0.9965 | 0.6546 |
|
| 72 |
+
| 0.1973 | 192.31 | 10000 | 1.0360 | 0.6551 |
|
| 73 |
+
|
| 74 |
+
|
| 75 |
+
### Framework versions
|
| 76 |
+
|
| 77 |
+
- Transformers 4.17.0.dev0
|
| 78 |
+
- Pytorch 1.10.2+cu102
|
| 79 |
+
- Datasets 1.18.2.dev0
|
| 80 |
+
- Tokenizers 0.11.0
|