metadata
license: cc-by-nc-4.0
base_model: facebook/timesformer-base-finetuned-k400
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: >-
timesformer-base-finetuned-k400-kinetic400-subset-epoch6-num_frame_10_myViT2window2_more_data_b4
results: []
timesformer-base-finetuned-k400-kinetic400-subset-epoch6-num_frame_10_myViT2window2_more_data_b4
This model is a fine-tuned version of facebook/timesformer-base-finetuned-k400 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1852
- Accuracy: 0.96
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: 72
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.4842 | 0.18 | 13 | 0.6124 | 0.92 |
| 0.3375 | 1.18 | 26 | 0.2410 | 0.96 |
| 0.0626 | 2.18 | 39 | 0.1339 | 1.0 |
| 0.0046 | 3.18 | 52 | 0.1199 | 0.96 |
| 0.0026 | 4.18 | 65 | 0.1185 | 0.96 |
| 0.0018 | 5.1 | 72 | 0.1166 | 0.96 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.19.0
- Tokenizers 0.15.2