|
--- |
|
library_name: transformers |
|
license: cc-by-nc-4.0 |
|
base_model: MCG-NJU/videomae-large-finetuned-kinetics |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: MAE-CT-CPC-Dicotomized-n0-m1-d1-v9 |
|
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. --> |
|
|
|
# MAE-CT-CPC-Dicotomized-n0-m1-d1-v9 |
|
|
|
This model is a fine-tuned version of [MCG-NJU/videomae-large-finetuned-kinetics](https://huggingface.co./MCG-NJU/videomae-large-finetuned-kinetics) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.6419 |
|
- Accuracy: 0.5 |
|
|
|
## 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: 1e-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: 2400 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-------:|:----:|:---------------:|:--------:| |
|
| 0.6819 | 0.0204 | 49 | 0.6641 | 0.5789 | |
|
| 0.6201 | 1.0204 | 98 | 0.6368 | 0.5789 | |
|
| 0.6787 | 2.0204 | 147 | 0.6109 | 0.5789 | |
|
| 0.6484 | 3.0204 | 196 | 0.6305 | 0.6316 | |
|
| 0.6357 | 4.0204 | 245 | 0.6794 | 0.5263 | |
|
| 0.5692 | 5.0204 | 294 | 0.6938 | 0.5263 | |
|
| 0.658 | 6.0204 | 343 | 0.6696 | 0.5789 | |
|
| 0.5112 | 7.0204 | 392 | 0.5669 | 0.6316 | |
|
| 0.3719 | 8.0204 | 441 | 0.8023 | 0.6316 | |
|
| 0.4573 | 9.0204 | 490 | 0.5991 | 0.6316 | |
|
| 0.2915 | 10.0204 | 539 | 0.7899 | 0.6842 | |
|
| 0.3723 | 11.0204 | 588 | 0.9848 | 0.5263 | |
|
| 0.1372 | 12.0204 | 637 | 0.9510 | 0.6316 | |
|
| 0.0066 | 13.0204 | 686 | 1.1872 | 0.5789 | |
|
| 0.2218 | 14.0204 | 735 | 1.1692 | 0.7368 | |
|
| 0.7666 | 15.0204 | 784 | 1.1125 | 0.6842 | |
|
| 0.2171 | 16.0204 | 833 | 1.2615 | 0.7368 | |
|
| 0.1152 | 17.0204 | 882 | 1.3683 | 0.7368 | |
|
| 0.1941 | 18.0204 | 931 | 1.3915 | 0.7368 | |
|
| 0.0233 | 19.0204 | 980 | 1.5972 | 0.5789 | |
|
| 0.2367 | 20.0204 | 1029 | 1.7445 | 0.6842 | |
|
| 0.1575 | 21.0204 | 1078 | 2.4036 | 0.6316 | |
|
| 0.0008 | 22.0204 | 1127 | 2.2633 | 0.6316 | |
|
| 0.019 | 23.0204 | 1176 | 2.2078 | 0.6842 | |
|
| 0.0006 | 24.0204 | 1225 | 2.1281 | 0.6316 | |
|
| 0.0004 | 25.0204 | 1274 | 2.0100 | 0.7368 | |
|
| 0.0002 | 26.0204 | 1323 | 2.3862 | 0.6316 | |
|
| 0.0003 | 27.0204 | 1372 | 2.1579 | 0.6842 | |
|
| 0.0372 | 28.0204 | 1421 | 2.1226 | 0.6842 | |
|
| 0.1547 | 29.0204 | 1470 | 2.4949 | 0.6842 | |
|
| 0.0015 | 30.0204 | 1519 | 1.8815 | 0.6842 | |
|
| 0.0005 | 31.0204 | 1568 | 2.2582 | 0.6316 | |
|
| 0.0002 | 32.0204 | 1617 | 2.1774 | 0.6316 | |
|
| 0.0001 | 33.0204 | 1666 | 2.2222 | 0.6316 | |
|
| 0.0001 | 34.0204 | 1715 | 2.3560 | 0.6316 | |
|
| 0.0001 | 35.0204 | 1764 | 2.3904 | 0.6316 | |
|
| 0.0001 | 36.0204 | 1813 | 2.2240 | 0.7368 | |
|
| 0.0002 | 37.0204 | 1862 | 2.2880 | 0.6842 | |
|
| 0.0002 | 38.0204 | 1911 | 2.3010 | 0.6842 | |
|
| 0.0001 | 39.0204 | 1960 | 2.3124 | 0.6842 | |
|
| 0.0001 | 40.0204 | 2009 | 2.2903 | 0.6842 | |
|
| 0.1744 | 41.0204 | 2058 | 2.3067 | 0.6842 | |
|
| 0.0001 | 42.0204 | 2107 | 2.3422 | 0.6842 | |
|
| 0.0001 | 43.0204 | 2156 | 2.3477 | 0.6842 | |
|
| 0.0001 | 44.0204 | 2205 | 2.3783 | 0.6316 | |
|
| 0.0001 | 45.0204 | 2254 | 2.3447 | 0.6842 | |
|
| 0.0001 | 46.0204 | 2303 | 2.3415 | 0.6842 | |
|
| 0.0001 | 47.0204 | 2352 | 2.3419 | 0.6842 | |
|
| 0.0001 | 48.02 | 2400 | 2.3470 | 0.6842 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.45.2 |
|
- Pytorch 2.0.1+cu117 |
|
- Datasets 3.0.1 |
|
- Tokenizers 0.20.0 |
|
|
