<!-- 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-finetuned-nba-2-class-8-batch-399-vid-multiclass-2
This model is a fine-tuned version of MCG-NJU/videomae-base-finetuned-kinetics on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2984
- F1: 0.875
- Accuracy: 0.875
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-06
- 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: 2450
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Accuracy |
|---|---|---|---|---|---|
| 0.6707 | 0.02 | 50 | 0.6794 | 0.5417 | 0.5417 |
| 0.6515 | 1.02 | 100 | 0.6673 | 0.6354 | 0.6354 |
| 0.6127 | 2.02 | 150 | 0.6359 | 0.6562 | 0.6562 |
| 0.5019 | 3.02 | 200 | 0.5753 | 0.6979 | 0.6979 |
| 0.3779 | 4.02 | 250 | 0.5218 | 0.7292 | 0.7292 |
| 0.5177 | 5.02 | 300 | 0.5115 | 0.7292 | 0.7292 |
| 0.2963 | 6.02 | 350 | 0.5388 | 0.7083 | 0.7083 |
| 0.3246 | 7.02 | 400 | 0.5649 | 0.7188 | 0.7188 |
| 0.3415 | 8.02 | 450 | 0.5980 | 0.75 | 0.75 |
| 0.3663 | 9.02 | 500 | 0.6346 | 0.7292 | 0.7292 |
| 0.1356 | 10.02 | 550 | 0.5965 | 0.7083 | 0.7083 |
| 0.1614 | 11.02 | 600 | 0.6816 | 0.7396 | 0.7396 |
| 0.3411 | 12.02 | 650 | 0.7246 | 0.7396 | 0.7396 |
| 0.2877 | 13.02 | 700 | 0.6570 | 0.7292 | 0.7292 |
| 0.303 | 14.02 | 750 | 0.7529 | 0.7396 | 0.7396 |
| 0.2492 | 15.02 | 800 | 0.8294 | 0.7083 | 0.7083 |
| 0.2397 | 16.02 | 850 | 0.9062 | 0.6875 | 0.6875 |
| 0.2039 | 17.02 | 900 | 0.6713 | 0.75 | 0.75 |
| 0.1591 | 18.02 | 950 | 0.9365 | 0.7188 | 0.7188 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.14.0
