File size: 3,015 Bytes
5b995b6 5dae949 5b995b6 5dae949 5b995b6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 |
---
license: mit
base_model: facebook/w2v-bert-2.0
tags:
- automatic-speech-recognition
- DewiBrynJones/banc-trawsgrifiadau-bangor-normalized
- generated_from_trainer
metrics:
- wer
model-index:
- name: w2v2-bert-ft-btb-cy
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# w2v2-bert-ft-btb-cy
This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-NORMALIZED - DEFAULT dataset.
It achieves the following results on the evaluation set:
- Loss: 2.9375
- Wer: 1.0
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 2500
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:----:|:---------------:|:---:|
| No log | 0.1414 | 100 | 6.9702 | 1.0 |
| No log | 0.2829 | 200 | 3.3676 | 1.0 |
| No log | 0.4243 | 300 | 3.0393 | 1.0 |
| No log | 0.5658 | 400 | 2.9887 | 1.0 |
| 4.717 | 0.7072 | 500 | 3.0227 | 1.0 |
| 4.717 | 0.8487 | 600 | 3.0406 | 1.0 |
| 4.717 | 0.9901 | 700 | 3.0029 | 1.0 |
| 4.717 | 1.1315 | 800 | 2.9483 | 1.0 |
| 4.717 | 1.2730 | 900 | 2.9511 | 1.0 |
| 3.0065 | 1.4144 | 1000 | 2.9473 | 1.0 |
| 3.0065 | 1.5559 | 1100 | 2.9448 | 1.0 |
| 3.0065 | 1.6973 | 1200 | 2.9470 | 1.0 |
| 3.0065 | 1.8388 | 1300 | 2.9446 | 1.0 |
| 3.0065 | 1.9802 | 1400 | 2.9432 | 1.0 |
| 2.9634 | 2.1216 | 1500 | 2.9476 | 1.0 |
| 2.9634 | 2.2631 | 1600 | 2.9624 | 1.0 |
| 2.9634 | 2.4045 | 1700 | 2.9581 | 1.0 |
| 2.9634 | 2.5460 | 1800 | 2.9553 | 1.0 |
| 2.9634 | 2.6874 | 1900 | 2.9515 | 1.0 |
| 2.9677 | 2.8289 | 2000 | 2.9481 | 1.0 |
| 2.9677 | 2.9703 | 2100 | 2.9509 | 1.0 |
| 2.9677 | 3.1117 | 2200 | 2.9408 | 1.0 |
| 2.9677 | 3.2532 | 2300 | 2.9393 | 1.0 |
| 2.9677 | 3.3946 | 2400 | 2.9381 | 1.0 |
| 2.9612 | 3.5361 | 2500 | 2.9375 | 1.0 |
### Framework versions
- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
|