metadata
license: mit
base_model: microsoft/git-base
tags:
- generated_from_trainer
model-index:
- name: git-base-pokemon
results: []
git-base-pokemon
This model is a fine-tuned version of microsoft/git-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0362
- Wer Score: 3.5483
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Score |
|---|---|---|---|---|
| 7.2786 | 4.17 | 50 | 4.4614 | 21.6293 |
| 2.2929 | 8.33 | 100 | 0.4321 | 12.5212 |
| 0.1294 | 12.5 | 150 | 0.0356 | 1.1338 |
| 0.0142 | 16.67 | 200 | 0.0332 | 4.6898 |
| 0.0039 | 20.83 | 250 | 0.0339 | 3.2767 |
| 0.0021 | 25.0 | 300 | 0.0349 | 3.9781 |
| 0.0016 | 29.17 | 350 | 0.0354 | 3.8867 |
| 0.0013 | 33.33 | 400 | 0.0359 | 3.6744 |
| 0.0012 | 37.5 | 450 | 0.0360 | 3.6564 |
| 0.0011 | 41.67 | 500 | 0.0363 | 3.4659 |
| 0.001 | 45.83 | 550 | 0.0363 | 3.5495 |
| 0.001 | 50.0 | 600 | 0.0362 | 3.5483 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.14.7
- Tokenizers 0.15.0