Human-action-convnext
This model is a fine-tuned version of facebook/convnext-tiny-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5382
- Accuracy: 0.8365
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 12
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.7559 | 0.99 | 78 | 0.6921 | 0.7940 |
0.6605 | 1.99 | 157 | 0.6305 | 0.8163 |
0.5673 | 3.0 | 236 | 0.5968 | 0.8179 |
0.4815 | 4.0 | 315 | 0.5834 | 0.8167 |
0.4497 | 4.99 | 393 | 0.5617 | 0.8294 |
0.4009 | 5.99 | 472 | 0.5505 | 0.8298 |
0.3469 | 7.0 | 551 | 0.5466 | 0.8333 |
0.3096 | 8.0 | 630 | 0.5417 | 0.8361 |
0.3083 | 8.99 | 708 | 0.5450 | 0.8313 |
0.333 | 9.99 | 787 | 0.5417 | 0.8345 |
0.2979 | 11.0 | 866 | 0.5386 | 0.8377 |
0.2469 | 11.89 | 936 | 0.5382 | 0.8365 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 191
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Model tree for luciagil/Human-action-convnext
Base model
facebook/convnext-tiny-224