metadata
library_name: transformers
license: cc-by-nc-4.0
base_model: MCG-NJU/videomae-large
tags:
- generated_from_trainer
model-index:
- name: videomae-large-finetuned-deepfake-subset
results: []
videomae-large-finetuned-deepfake-subset
This model is a fine-tuned version of MCG-NJU/videomae-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3932
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 4470
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6361 | 0.1 | 447 | 0.6478 |
| 0.6774 | 1.1 | 894 | 0.6047 |
| 0.4168 | 2.1 | 1341 | 0.4852 |
| 0.4427 | 3.1 | 1788 | 0.8547 |
| 0.4496 | 4.1 | 2235 | 0.3795 |
| 0.3433 | 5.1 | 2682 | 0.4119 |
| 0.2287 | 6.1 | 3129 | 0.4823 |
| 0.1297 | 7.1 | 3576 | 0.4295 |
| 0.3104 | 8.1 | 4023 | 0.4096 |
| 0.0525 | 9.1 | 4470 | 0.4154 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Datasets 3.1.0
- Tokenizers 0.19.1