Vivit-d3
This model is a fine-tuned version of google/vivit-b-16x2-kinetics400 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2800
- Accuracy: 0.9509
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: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 2240
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 3.113 | 0.1 | 224 | 0.2148 | 0.9464 |
| 3.4763 | 1.1 | 448 | 0.3339 | 0.8869 |
| 0.5019 | 2.1 | 672 | 0.3398 | 0.9449 |
| 0.0116 | 3.1 | 896 | 0.3553 | 0.9360 |
| 0.8144 | 4.1 | 1120 | 0.4592 | 0.9405 |
| 0.9856 | 5.1 | 1344 | 0.3184 | 0.9286 |
| 0.0005 | 6.1 | 1568 | 0.2253 | 0.9435 |
| 0.0001 | 7.1 | 1792 | 0.3713 | 0.9479 |
| 0.0 | 8.1 | 2016 | 0.3450 | 0.9479 |
| 0.2266 | 9.1 | 2240 | 0.2800 | 0.9509 |
Framework versions
- Transformers 4.46.2
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.20.3