my_awesome_pokemon_model_resnet18
This model is a fine-tuned version of microsoft/resnet-18 on the pokemon-classification dataset. It achieves the following results on the evaluation set:
- Loss: 6.8019
- Accuracy: 0.0108
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: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 4.275 | 1.0 | 76 | 6.1680 | 0.0014 |
| 3.3896 | 1.99 | 152 | 6.6421 | 0.0115 |
| 3.0563 | 2.99 | 228 | 6.8019 | 0.0108 |
Framework versions
- Transformers 4.29.1
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.3