metadata
license: mit
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: roberta-gest-pred-seqeval-partialmatch
results: []
datasets:
- Jsevisal/gesture_pred
roberta-gest-pred-seqeval-partialmatch
This model is a fine-tuned version of roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.624567
- Precision: 0.802308
- Recall: 0.738156
- F1: 0.744604
- Accuracy: 0.834450
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: 2e-05
- 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 | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 2.0269 | 1.0 | 147 | 1.2260 | 0.3446 | 0.3571 | 0.3404 | 0.6696 |
| 1.0422 | 2.0 | 294 | 0.8553 | 0.5596 | 0.5248 | 0.4885 | 0.7594 |
| 0.7198 | 3.0 | 441 | 0.7086 | 0.6623 | 0.6298 | 0.6110 | 0.8097 |
| 0.5231 | 4.0 | 588 | 0.6330 | 0.7415 | 0.7102 | 0.7061 | 0.8264 |
| 0.3947 | 5.0 | 735 | 0.6246 | 0.8023 | 0.7382 | 0.7446 | 0.8345 |
| 0.2866 | 6.0 | 882 | 0.6487 | 0.8263 | 0.7578 | 0.7496 | 0.8519 |
| 0.2338 | 7.0 | 1029 | 0.6662 | 0.7970 | 0.7608 | 0.7452 | 0.8465 |
| 0.1791 | 8.0 | 1176 | 0.6762 | 0.7923 | 0.7690 | 0.7432 | 0.8398 |
| 0.1495 | 9.0 | 1323 | 0.6496 | 0.8008 | 0.7946 | 0.7686 | 0.8552 |
| 0.1316 | 10.0 | 1470 | 0.6909 | 0.7952 | 0.7778 | 0.7489 | 0.8458 |
Framework versions
- Transformers 4.27.3
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2