update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: wav2vec2-large-xls-r-1b-bemba-fds
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# wav2vec2-large-xls-r-1b-bemba-fds
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 0.2898
|
| 18 |
+
- Wer: 0.3435
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 5e-05
|
| 38 |
+
- train_batch_size: 4
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- gradient_accumulation_steps: 2
|
| 42 |
+
- total_train_batch_size: 8
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- lr_scheduler_warmup_steps: 500
|
| 46 |
+
- num_epochs: 15
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 53 |
+
| 1.7986 | 0.34 | 500 | 0.4549 | 0.7292 |
|
| 54 |
+
| 0.5358 | 0.67 | 1000 | 0.3325 | 0.4491 |
|
| 55 |
+
| 0.4559 | 1.01 | 1500 | 0.3090 | 0.3954 |
|
| 56 |
+
| 0.3983 | 1.35 | 2000 | 0.3067 | 0.4105 |
|
| 57 |
+
| 0.4067 | 1.68 | 2500 | 0.2838 | 0.3678 |
|
| 58 |
+
| 0.3722 | 2.02 | 3000 | 0.2824 | 0.3762 |
|
| 59 |
+
| 0.3286 | 2.36 | 3500 | 0.2810 | 0.3670 |
|
| 60 |
+
| 0.3239 | 2.69 | 4000 | 0.2643 | 0.3501 |
|
| 61 |
+
| 0.3187 | 3.03 | 4500 | 0.2838 | 0.3754 |
|
| 62 |
+
| 0.2801 | 3.36 | 5000 | 0.2815 | 0.3507 |
|
| 63 |
+
| 0.2806 | 3.7 | 5500 | 0.2725 | 0.3486 |
|
| 64 |
+
| 0.2714 | 4.04 | 6000 | 0.2898 | 0.3435 |
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
### Framework versions
|
| 68 |
+
|
| 69 |
+
- Transformers 4.16.2
|
| 70 |
+
- Pytorch 1.10.0+cu111
|
| 71 |
+
- Datasets 1.18.3
|
| 72 |
+
- Tokenizers 0.11.0
|