metadata
license: cc-by-nc-4.0
base_model: facebook/timesformer-base-finetuned-k600
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: model_timesformer_subset_02
results: []
model_timesformer_subset_02
This model is a fine-tuned version of facebook/timesformer-base-finetuned-k600 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.7125
- Accuracy: 0.5446
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
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 2997
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.8295 | 0.08 | 250 | 1.9005 | 0.4684 |
| 1.6577 | 1.08 | 500 | 1.7229 | 0.4908 |
| 1.4384 | 2.08 | 750 | 1.6627 | 0.5172 |
| 1.2867 | 3.08 | 1000 | 1.7071 | 0.5182 |
| 1.0143 | 4.08 | 1250 | 1.6543 | 0.5381 |
| 0.8775 | 5.08 | 1500 | 1.6755 | 0.5286 |
| 0.9622 | 6.08 | 1750 | 1.6568 | 0.5401 |
| 0.5627 | 7.08 | 2000 | 1.6966 | 0.5371 |
| 0.6586 | 8.08 | 2250 | 1.6735 | 0.5531 |
| 0.4577 | 9.08 | 2500 | 1.7446 | 0.5267 |
| 0.4552 | 10.08 | 2750 | 1.7490 | 0.5371 |
| 0.4488 | 11.08 | 2997 | 1.7125 | 0.5446 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.1.2
- Datasets 2.1.0
- Tokenizers 0.15.2