update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: wav2vec2-xlsr-fi-lm-1B
|
| 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-xlsr-fi-lm-1B
|
| 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: 0.1853
|
| 18 |
+
- Wer: 0.2205
|
| 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: 0.0003
|
| 38 |
+
- train_batch_size: 8
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
+
- seed: 42
|
| 41 |
+
- gradient_accumulation_steps: 4
|
| 42 |
+
- total_train_batch_size: 32
|
| 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: 10
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 53 |
+
| 0.8158 | 0.67 | 400 | 0.4835 | 0.6310 |
|
| 54 |
+
| 0.5679 | 1.33 | 800 | 0.4806 | 0.5538 |
|
| 55 |
+
| 0.6055 | 2.0 | 1200 | 0.3888 | 0.5083 |
|
| 56 |
+
| 0.5353 | 2.67 | 1600 | 0.3258 | 0.4365 |
|
| 57 |
+
| 0.4883 | 3.33 | 2000 | 0.3313 | 0.4204 |
|
| 58 |
+
| 0.4513 | 4.0 | 2400 | 0.2924 | 0.3904 |
|
| 59 |
+
| 0.3753 | 4.67 | 2800 | 0.2593 | 0.3608 |
|
| 60 |
+
| 0.3478 | 5.33 | 3200 | 0.2832 | 0.3551 |
|
| 61 |
+
| 0.3796 | 6.0 | 3600 | 0.2495 | 0.3402 |
|
| 62 |
+
| 0.2556 | 6.67 | 4000 | 0.2342 | 0.3106 |
|
| 63 |
+
| 0.229 | 7.33 | 4400 | 0.2181 | 0.2812 |
|
| 64 |
+
| 0.205 | 8.0 | 4800 | 0.2041 | 0.2523 |
|
| 65 |
+
| 0.1654 | 8.67 | 5200 | 0.2015 | 0.2416 |
|
| 66 |
+
| 0.152 | 9.33 | 5600 | 0.1942 | 0.2294 |
|
| 67 |
+
| 0.1569 | 10.0 | 6000 | 0.1853 | 0.2205 |
|
| 68 |
+
|
| 69 |
+
|
| 70 |
+
### Framework versions
|
| 71 |
+
|
| 72 |
+
- Transformers 4.16.0.dev0
|
| 73 |
+
- Pytorch 1.10.1+cu102
|
| 74 |
+
- Datasets 1.17.1.dev0
|
| 75 |
+
- Tokenizers 0.11.0
|