metadata
license: cc-by-nc-4.0
base_model: MCG-NJU/videomae-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: videomae-base-finetuned-good-gesturePhaseV1
results: []
videomae-base-finetuned-good-gesturePhaseV1
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:
- Accuracy: 0.5918
- Loss: 1.1628
- Accuracy Hold: 0.0
- Accuracy Stroke: 0.0
- Accuracy Recovery: 0.0
- Accuracy Preparation: 0.9221
- Accuracy Unknown: 0.4571
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: 8
- eval_batch_size: 8
- 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: 276
Training results
| Training Loss | Epoch | Step | Accuracy | Validation Loss | Accuracy Hold | Accuracy Stroke | Accuracy Recovery | Accuracy Preparation | Accuracy Unknown |
|---|---|---|---|---|---|---|---|---|---|
| 1.2297 | 0.2536 | 70 | 0.5597 | 1.2679 | 0.0 | 0.0 | 0.0 | 1.0 | 0.0 |
| 0.9724 | 1.2536 | 140 | 0.5597 | 1.2204 | 0.0 | 0.0 | 0.0 | 1.0 | 0.0 |
| 0.9151 | 2.2536 | 210 | 0.6119 | 1.0776 | 0.0 | 0.0 | 0.0 | 0.9333 | 0.4615 |
| 0.9169 | 3.2391 | 276 | 0.6269 | 1.0385 | 0.0 | 0.0 | 0.0 | 0.9067 | 0.6154 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1