|
--- |
|
license: cc-by-nc-4.0 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: videomae-base-finetuned-ucf101-subset |
|
results: [] |
|
--- |
|
|
|
<!-- 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-ucf101-subset |
|
|
|
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1834 |
|
- Accuracy: 0.9290 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 4 |
|
- 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: 3750 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 2.2357 | 0.02 | 75 | 2.1952 | 0.3429 | |
|
| 1.8271 | 1.02 | 150 | 1.8570 | 0.3571 | |
|
| 0.8947 | 2.02 | 225 | 0.8397 | 0.7286 | |
|
| 0.5347 | 3.02 | 300 | 0.5632 | 0.8286 | |
|
| 0.4101 | 4.02 | 375 | 0.6908 | 0.8 | |
|
| 0.0855 | 5.02 | 450 | 0.1541 | 0.9571 | |
|
| 0.2286 | 6.02 | 525 | 0.2762 | 0.9 | |
|
| 0.2905 | 7.02 | 600 | 0.2306 | 0.9 | |
|
| 0.0051 | 8.02 | 675 | 0.2054 | 0.9571 | |
|
| 0.1142 | 9.02 | 750 | 0.7049 | 0.8714 | |
|
| 0.0022 | 10.02 | 825 | 0.1919 | 0.9429 | |
|
| 0.0019 | 11.02 | 900 | 0.5478 | 0.8857 | |
|
| 0.0021 | 12.02 | 975 | 0.4232 | 0.9286 | |
|
| 0.0019 | 13.02 | 1050 | 0.5437 | 0.8857 | |
|
| 0.002 | 14.02 | 1125 | 1.0354 | 0.7857 | |
|
| 0.0028 | 15.02 | 1200 | 0.1039 | 0.9714 | |
|
| 0.0013 | 16.02 | 1275 | 0.1552 | 0.9714 | |
|
| 0.2309 | 17.02 | 1350 | 0.2720 | 0.9286 | |
|
| 0.1662 | 18.02 | 1425 | 0.0312 | 0.9857 | |
|
| 0.0199 | 19.02 | 1500 | 0.1478 | 0.9571 | |
|
| 0.001 | 20.02 | 1575 | 0.2189 | 0.9714 | |
|
| 0.0009 | 21.02 | 1650 | 0.1568 | 0.9714 | |
|
| 0.0008 | 22.02 | 1725 | 0.2136 | 0.9429 | |
|
| 0.0007 | 23.02 | 1800 | 0.1032 | 0.9714 | |
|
| 0.0007 | 24.02 | 1875 | 0.1026 | 0.9714 | |
|
| 0.0006 | 25.02 | 1950 | 0.1130 | 0.9571 | |
|
| 0.0006 | 26.02 | 2025 | 0.1147 | 0.9714 | |
|
| 0.0006 | 27.02 | 2100 | 0.0858 | 0.9857 | |
|
| 0.0006 | 28.02 | 2175 | 0.0868 | 0.9857 | |
|
| 0.0006 | 29.02 | 2250 | 0.0880 | 0.9857 | |
|
| 0.0005 | 30.02 | 2325 | 0.0872 | 0.9857 | |
|
| 0.0005 | 31.02 | 2400 | 0.0901 | 0.9857 | |
|
| 0.0005 | 32.02 | 2475 | 0.0894 | 0.9857 | |
|
| 0.0005 | 33.02 | 2550 | 0.0858 | 0.9857 | |
|
| 0.0005 | 34.02 | 2625 | 0.0912 | 0.9857 | |
|
| 0.0005 | 35.02 | 2700 | 0.3303 | 0.9429 | |
|
| 0.0004 | 36.02 | 2775 | 0.1737 | 0.9429 | |
|
| 0.0004 | 37.02 | 2850 | 0.1534 | 0.9714 | |
|
| 0.0004 | 38.02 | 2925 | 0.1350 | 0.9714 | |
|
| 0.0004 | 39.02 | 3000 | 0.1270 | 0.9714 | |
|
| 0.0004 | 40.02 | 3075 | 0.1253 | 0.9714 | |
|
| 0.0004 | 41.02 | 3150 | 0.1241 | 0.9714 | |
|
| 0.0004 | 42.02 | 3225 | 0.1247 | 0.9714 | |
|
| 0.0004 | 43.02 | 3300 | 0.1262 | 0.9714 | |
|
| 0.0004 | 44.02 | 3375 | 0.1266 | 0.9571 | |
|
| 0.0004 | 45.02 | 3450 | 0.1741 | 0.9714 | |
|
| 0.0004 | 46.02 | 3525 | 0.1753 | 0.9714 | |
|
| 0.0004 | 47.02 | 3600 | 0.1634 | 0.9714 | |
|
| 0.0004 | 48.02 | 3675 | 0.1603 | 0.9714 | |
|
| 0.0004 | 49.02 | 3750 | 0.1602 | 0.9714 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.24.0 |
|
- Pytorch 1.8.0+cu111 |
|
- Datasets 2.7.1 |
|
- Tokenizers 0.13.2 |
|
|
