<!-- 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-5-class-4-batch-8000-vid-multilabel-3
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: 0.3584
- F1: 0.8505
- Roc Auc: 0.8870
- Accuracy: 0.7402
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: 1.5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 50000
Training results
Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
---|---|---|---|---|---|---|
0.4913 | 0.04 | 2000 | 0.5028 | 0.5811 | 0.7020 | 0.2147 |
0.296 | 1.04 | 4000 | 0.4175 | 0.6805 | 0.7646 | 0.3674 |
0.3384 | 2.04 | 6000 | 0.4054 | 0.6912 | 0.7715 | 0.3821 |
0.3212 | 3.04 | 8000 | 0.4071 | 0.7029 | 0.7762 | 0.4958 |
0.2955 | 4.04 | 10000 | 0.3978 | 0.7144 | 0.7853 | 0.5158 |
0.2713 | 5.04 | 12000 | 0.3738 | 0.7550 | 0.8113 | 0.5758 |
0.2448 | 6.04 | 14000 | 0.3673 | 0.7663 | 0.8189 | 0.5642 |
0.3435 | 7.04 | 16000 | 0.3496 | 0.7960 | 0.8404 | 0.6284 |
0.2489 | 8.04 | 18000 | 0.3656 | 0.7970 | 0.8426 | 0.6526 |
0.2794 | 9.04 | 20000 | 0.3582 | 0.7972 | 0.8428 | 0.64 |
0.36 | 10.04 | 22000 | 0.3313 | 0.8223 | 0.8611 | 0.6811 |
0.2545 | 11.04 | 24000 | 0.3392 | 0.8408 | 0.8771 | 0.7168 |
0.2076 | 12.04 | 26000 | 0.3582 | 0.8254 | 0.8698 | 0.6979 |
0.2294 | 13.04 | 28000 | 0.3469 | 0.8238 | 0.8630 | 0.7032 |
0.1248 | 14.04 | 30000 | 0.3482 | 0.8432 | 0.8780 | 0.7379 |
0.153 | 15.04 | 32000 | 0.3392 | 0.8477 | 0.8795 | 0.7274 |
0.1764 | 16.04 | 34000 | 0.3410 | 0.8395 | 0.8731 | 0.7295 |
0.2392 | 17.04 | 36000 | 0.3300 | 0.8612 | 0.8912 | 0.7621 |
0.2256 | 18.04 | 38000 | 0.3344 | 0.8570 | 0.8892 | 0.7558 |
0.2725 | 19.04 | 40000 | 0.3286 | 0.8604 | 0.8909 | 0.7653 |
0.1801 | 20.04 | 42000 | 0.3439 | 0.8626 | 0.8939 | 0.7684 |
0.2238 | 21.04 | 44000 | 0.3446 | 0.8564 | 0.8894 | 0.7621 |
0.2488 | 22.04 | 46000 | 0.3491 | 0.8643 | 0.8946 | 0.78 |
0.0829 | 23.04 | 48000 | 0.3464 | 0.8625 | 0.8934 | 0.7684 |
0.1616 | 24.04 | 50000 | 0.3475 | 0.8599 | 0.8921 | 0.7716 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1