metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- image-classification
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: vit-Valence
results: []
Image-Valence
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the custom dataset. It achieves the following results on the evaluation set:
- Loss: 1.4464
- Accuracy: 0.5863
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.2256 | 0.78 | 100 | 1.0936 | 0.5451 |
| 0.7315 | 1.56 | 200 | 0.9981 | 0.5882 |
| 0.2118 | 2.34 | 300 | 1.1650 | 0.5902 |
| 0.1119 | 3.12 | 400 | 1.2864 | 0.5863 |
| 0.1116 | 3.91 | 500 | 1.4464 | 0.5863 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.1