roberta-finetuned-gesture-prediction-5-classes

This model is a fine-tuned version of roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4764
• Accuracy: 0.8729
• Precision: 0.8731
• Recall: 0.8729
• F1: 0.8725

It achieves the following results on the evaluation set:

• Loss: 0.4842
• Accuracy: 0.8628
• Precision: 0.8629
• Recall: 0.8628
• F1: 0.8619

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

The model has been trained with the qfrodicio/gesture-prediction-5-classes dataset

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	Accuracy	Precision	Recall	F1
1.4556	1.0	71	0.9405	0.6561	0.6129	0.6561	0.5981
0.7207	2.0	142	0.5276	0.8442	0.8463	0.8442	0.8406
0.4005	3.0	213	0.4997	0.8662	0.8719	0.8662	0.8640
0.2417	4.0	284	0.4764	0.8729	0.8731	0.8729	0.8725
0.1757	5.0	355	0.5135	0.8812	0.8827	0.8812	0.8810
0.1398	6.0	426	0.5266	0.8710	0.8710	0.8710	0.8704
0.0937	7.0	497	0.5438	0.8799	0.8801	0.8799	0.8792
0.07	8.0	568	0.5759	0.8769	0.8770	0.8769	0.8766
0.0552	9.0	639	0.6035	0.8745	0.8741	0.8745	0.8738
0.0478	10.0	710	0.5974	0.8778	0.8775	0.8778	0.8771

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2