<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
distilbert-finetuned-gesture-prediction-21-classes
This model is a fine-tuned version of distilbert-base-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8430
- Accuracy: 0.8077
- Precision: 0.8063
- Recall: 0.8077
- F1: 0.8038
It achieves the following results on the test set:
- Loss: 0.8332
- Accuracy: 0.7934
- Precision: 0.7925
- Recall: 0.7934
- F1: 0.7875
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
This model has been trained with the qfrodicio/gesture-prediction-21-classes dataset
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- weight_decay: 0.01
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 2.2082 | 1.0 | 104 | 1.3318 | 0.6956 | 0.6361 | 0.6956 | 0.6473 |
| 1.1512 | 2.0 | 208 | 1.0114 | 0.7604 | 0.7463 | 0.7604 | 0.7368 |
| 0.8152 | 3.0 | 312 | 0.8805 | 0.7860 | 0.7677 | 0.7860 | 0.7698 |
| 0.6142 | 4.0 | 416 | 0.8486 | 0.8025 | 0.8035 | 0.8025 | 0.7961 |
| 0.4726 | 5.0 | 520 | 0.8651 | 0.7992 | 0.7987 | 0.7992 | 0.7894 |
| 0.3677 | 6.0 | 624 | 0.8430 | 0.8077 | 0.8063 | 0.8077 | 0.8038 |
| 0.2967 | 7.0 | 728 | 0.8564 | 0.8037 | 0.8029 | 0.8037 | 0.7995 |
| 0.2494 | 8.0 | 832 | 0.8567 | 0.8077 | 0.8054 | 0.8077 | 0.8041 |
| 0.2163 | 9.0 | 936 | 0.8789 | 0.8075 | 0.8060 | 0.8075 | 0.8035 |
| 0.193 | 10.0 | 1040 | 0.8880 | 0.8077 | 0.8072 | 0.8077 | 0.8032 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2
