videomae-small-finetuned-kinetics-finetuned-xd-violence
This model is a fine-tuned version of MCG-NJU/videomae-small-finetuned-kinetics on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3828
- Accuracy: 0.8254
Video preprocessing config: Resize to: (224, 224) Frames to sample: 16 Clip duration: 2.1333333333333333s
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.3574 | 1.0 | 394 | 0.5181 | 0.7738 |
| 0.4428 | 2.0 | 788 | 0.4357 | 0.8119 |
| 0.075 | 3.0 | 1182 | 0.4100 | 0.8183 |
Testing results
{'eval_loss': 0.3827999532222748,
'eval_accuracy': 0.8253768844221105,
'eval_runtime': 184.3164,
'eval_samples_per_second': 4.319,
'eval_steps_per_second': 2.159,
'epoch': 3.0}
| Class | Precision | Recall | F1-Score | Support |
|---|---|---|---|---|
| Normal (0) | 0.74 | 0.83 | 0.78 | 299 |
| Violent (1) | 0.89 | 0.82 | 0.85 | 497 |
| Accuracy | 0.83 | 796 | ||
| Macro Avg | 0.81 | 0.83 | 0.82 | 796 |
| Weighted Avg | 0.83 | 0.83 | 0.83 | 796 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1