metadata
license: cc-by-nc-4.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: videomae-base-finetuned-ucf101-subset
results: []
videomae-base-finetuned-ucf101-subset
This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2613
- Accuracy: 0.8968
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: 375
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.8376 | 0.2 | 75 | 1.3921 | 0.5571 |
| 0.569 | 1.2 | 150 | 0.6125 | 0.7714 |
| 0.32 | 2.2 | 225 | 0.4814 | 0.7857 |
| 0.2384 | 3.2 | 300 | 0.3162 | 0.9 |
| 0.0338 | 4.2 | 375 | 0.1798 | 0.9429 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.10.0
- Datasets 2.7.1
- Tokenizers 0.12.1