Rebecca Iglesias-Flores
commited on
Commit
•
1f5751e
1
Parent(s):
e6eb03c
End of training
Browse files
README.md
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: facebook/wav2vec2-xls-r-300m
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- wer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: finetuned_Wav2Vec2_on_ATCOSIM
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# finetuned_Wav2Vec2_on_ATCOSIM
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.1707
|
| 21 |
+
- Wer: 0.1170
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0003
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- gradient_accumulation_steps: 5
|
| 45 |
+
- total_train_batch_size: 40
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: linear
|
| 48 |
+
- lr_scheduler_warmup_steps: 500
|
| 49 |
+
- num_epochs: 30
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 55 |
+
| 2.6927 | 2.5 | 400 | 0.5557 | 0.4100 |
|
| 56 |
+
| 0.3288 | 4.99 | 800 | 0.2382 | 0.1943 |
|
| 57 |
+
| 0.1856 | 7.49 | 1200 | 0.1957 | 0.1699 |
|
| 58 |
+
| 0.1325 | 9.99 | 1600 | 0.1845 | 0.1572 |
|
| 59 |
+
| 0.1018 | 12.48 | 2000 | 0.1771 | 0.1534 |
|
| 60 |
+
| 0.0899 | 14.98 | 2400 | 0.1637 | 0.1356 |
|
| 61 |
+
| 0.0722 | 17.48 | 2800 | 0.1812 | 0.1409 |
|
| 62 |
+
| 0.0596 | 19.98 | 3200 | 0.1747 | 0.1323 |
|
| 63 |
+
| 0.046 | 22.47 | 3600 | 0.1505 | 0.1307 |
|
| 64 |
+
| 0.037 | 24.97 | 4000 | 0.1705 | 0.1224 |
|
| 65 |
+
| 0.0294 | 27.47 | 4400 | 0.1614 | 0.1164 |
|
| 66 |
+
| 0.0249 | 29.96 | 4800 | 0.1707 | 0.1170 |
|
| 67 |
+
|
| 68 |
+
|
| 69 |
+
### Framework versions
|
| 70 |
+
|
| 71 |
+
- Transformers 4.34.0.dev0
|
| 72 |
+
- Pytorch 2.0.1+cu118
|
| 73 |
+
- Datasets 2.14.5
|
| 74 |
+
- Tokenizers 0.13.3
|