metadata
license: mit
base_model: microsoft/git-base
tags:
- generated_from_trainer
datasets:
- imagefolder
model-index:
- name: git-base-pokemon
results: []
git-base-pokemon
This model is a fine-tuned version of microsoft/git-base on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0330
- Wer Score: 1.6516
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: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Score |
|---|---|---|---|---|
| 7.4022 | 4.17 | 50 | 4.7553 | 21.1384 |
| 2.7988 | 8.33 | 100 | 0.9177 | 10.7623 |
| 0.3496 | 12.5 | 150 | 0.0709 | 2.1170 |
| 0.0373 | 16.67 | 200 | 0.0327 | 1.3170 |
| 0.0142 | 20.83 | 250 | 0.0316 | 1.5031 |
| 0.0069 | 25.0 | 300 | 0.0330 | 1.6516 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3