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-finetuned-kinetic-subset-three-local-temporal-with-spatial
results: []
timesformer-base-finetuned-k400-finetuned-kinetic-subset-three-local-temporal-with-spatial
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: 0.3731
- Accuracy: 0.9118
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: 1
- eval_batch_size: 1
- 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: 300
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.0003 | 0.17 | 50 | 0.4553 | 0.8472 |
| 0.0004 | 1.17 | 100 | 0.3349 | 0.9028 |
| 0.0004 | 2.17 | 150 | 0.1279 | 0.9583 |
| 0.0995 | 3.17 | 200 | 0.1181 | 0.9306 |
| 0.0001 | 4.17 | 250 | 0.1797 | 0.9167 |
| 0.0003 | 5.17 | 300 | 0.1393 | 0.9444 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2