bert-finetuned-gesture-prediction-21-classes

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

Loss: 0.8664
Accuracy: 0.8123
Precision: 0.8122
Recall: 0.8123
F1: 0.8048

It achieves the following results on the test set:

Loss: 0.8381
Accuracy: 0.7884
Precision: 0.7954
Recall: 0.7884
F1: 0.7827

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-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
2.225	1.0	104	1.3314	0.7115	0.6469	0.7115	0.6675
1.0881	2.0	208	0.9569	0.7750	0.7577	0.7750	0.7525
0.7006	3.0	312	0.8805	0.7959	0.7917	0.7959	0.7831
0.4943	4.0	416	0.8664	0.8123	0.8122	0.8123	0.8048
0.3372	5.0	520	0.8765	0.8130	0.8102	0.8130	0.8053
0.2416	6.0	624	0.8772	0.8166	0.8139	0.8166	0.8107
0.178	7.0	728	0.9186	0.8217	0.8186	0.8217	0.8167
0.1302	8.0	832	0.9186	0.8202	0.8183	0.8202	0.8165
0.1063	9.0	936	0.9618	0.8245	0.8213	0.8245	0.8198
0.094	10.0	1040	0.9660	0.8214	0.8184	0.8214	0.8166

Framework versions

Transformers 4.26.1
Pytorch 1.13.1+cu116
Datasets 2.10.1
Tokenizers 0.13.2