<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
videomae-base-finetuned-ASBD
This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7908
- Accuracy: 0.6912
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: 12
- eval_batch_size: 12
- 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: 500
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.274 | 0.1 | 49 | 1.3780 | 0.1892 |
| 1.2578 | 1.1 | 98 | 1.4805 | 0.3378 |
| 0.8274 | 2.1 | 147 | 1.0223 | 0.5270 |
| 0.6943 | 3.1 | 196 | 1.1050 | 0.6486 |
| 0.4192 | 4.1 | 245 | 1.3384 | 0.6081 |
| 0.3951 | 5.1 | 294 | 1.6994 | 0.5541 |
| 0.2655 | 6.1 | 343 | 1.7021 | 0.5946 |
| 0.0256 | 7.1 | 392 | 1.5357 | 0.6216 |
| 0.0368 | 8.1 | 441 | 1.6924 | 0.6892 |
| 0.0275 | 9.1 | 490 | 1.6663 | 0.6892 |
| 0.0364 | 10.02 | 500 | 1.6649 | 0.6757 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
