|
--- |
|
license: cc-by-nc-4.0 |
|
base_model: MCG-NJU/videomae-large |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- precision |
|
- recall |
|
model-index: |
|
- name: videomae-large_5class_UCFCrime |
|
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-large_5class_UCFCrime |
|
|
|
This model is a fine-tuned version of [MCG-NJU/videomae-large](https://huggingface.co/MCG-NJU/videomae-large) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.6826 |
|
- Accuracy: 0.7523 |
|
- Precision: 0.7853 |
|
- Recall: 0.7523 |
|
- F1 Score: 0.7364 |
|
|
|
## 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: 2 |
|
- eval_batch_size: 2 |
|
- 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: 2850 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 Score | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:--------:| |
|
| 1.5794 | 0.03 | 95 | 1.4947 | 0.4128 | 0.2370 | 0.4128 | 0.2975 | |
|
| 1.7243 | 1.03 | 190 | 1.3362 | 0.5321 | 0.4612 | 0.5321 | 0.4143 | |
|
| 1.6818 | 2.03 | 285 | 1.3317 | 0.3119 | 0.4356 | 0.3119 | 0.2364 | |
|
| 1.0727 | 3.03 | 380 | 1.1666 | 0.5046 | 0.5546 | 0.5046 | 0.4337 | |
|
| 1.1121 | 4.03 | 475 | 1.3545 | 0.4587 | 0.6563 | 0.4587 | 0.3678 | |
|
| 1.0062 | 5.03 | 570 | 1.4656 | 0.5963 | 0.5376 | 0.5963 | 0.5325 | |
|
| 1.2532 | 6.03 | 665 | 1.6516 | 0.5780 | 0.5596 | 0.5780 | 0.4982 | |
|
| 1.9184 | 7.03 | 760 | 1.5020 | 0.5872 | 0.7283 | 0.5872 | 0.5449 | |
|
| 1.0223 | 8.03 | 855 | 1.4417 | 0.5872 | 0.5636 | 0.5872 | 0.5027 | |
|
| 0.9406 | 9.03 | 950 | 1.9402 | 0.5780 | 0.5800 | 0.5780 | 0.4906 | |
|
| 1.3058 | 10.03 | 1045 | 1.7611 | 0.4679 | 0.6914 | 0.4679 | 0.4463 | |
|
| 0.9196 | 11.03 | 1140 | 1.0373 | 0.6330 | 0.5648 | 0.6330 | 0.5625 | |
|
| 0.4191 | 12.03 | 1235 | 0.9139 | 0.6789 | 0.7553 | 0.6789 | 0.6824 | |
|
| 0.4816 | 13.03 | 1330 | 1.0840 | 0.7248 | 0.8001 | 0.7248 | 0.7176 | |
|
| 1.0577 | 14.03 | 1425 | 0.9822 | 0.7339 | 0.7724 | 0.7339 | 0.7216 | |
|
| 0.719 | 15.03 | 1520 | 1.4597 | 0.6789 | 0.6954 | 0.6789 | 0.6715 | |
|
| 0.3427 | 16.03 | 1615 | 1.4807 | 0.6789 | 0.7114 | 0.6789 | 0.6818 | |
|
| 0.6303 | 17.03 | 1710 | 1.9664 | 0.6881 | 0.7482 | 0.6881 | 0.6604 | |
|
| 0.0025 | 18.03 | 1805 | 1.5750 | 0.7156 | 0.7403 | 0.7156 | 0.7067 | |
|
| 0.2404 | 19.03 | 1900 | 2.2045 | 0.6606 | 0.7090 | 0.6606 | 0.6292 | |
|
| 0.0313 | 20.03 | 1995 | 1.6007 | 0.7248 | 0.7776 | 0.7248 | 0.7091 | |
|
| 0.3372 | 21.03 | 2090 | 1.6536 | 0.7156 | 0.7811 | 0.7156 | 0.6864 | |
|
| 0.5431 | 22.03 | 2185 | 2.1961 | 0.6514 | 0.7559 | 0.6514 | 0.6161 | |
|
| 0.0003 | 23.03 | 2280 | 1.6826 | 0.7523 | 0.7853 | 0.7523 | 0.7364 | |
|
| 0.0013 | 24.03 | 2375 | 1.8359 | 0.7431 | 0.7906 | 0.7431 | 0.7291 | |
|
| 0.0025 | 25.03 | 2470 | 2.0891 | 0.6881 | 0.7433 | 0.6881 | 0.6611 | |
|
| 0.0153 | 26.03 | 2565 | 1.8442 | 0.7431 | 0.7855 | 0.7431 | 0.7296 | |
|
| 0.0147 | 27.03 | 2660 | 1.8483 | 0.7431 | 0.7816 | 0.7431 | 0.7243 | |
|
| 0.0002 | 28.03 | 2755 | 1.9301 | 0.7339 | 0.7765 | 0.7339 | 0.7150 | |
|
| 0.0004 | 29.03 | 2850 | 1.9169 | 0.7339 | 0.7765 | 0.7339 | 0.7150 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
