raffaelsiregar/rgai-emotions-classification

results

This model is a fine-tuned version of google/vit-base-patch32-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.5006
• Accuracy: 0.4813

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.0441	1.0	40	2.0365	0.25
1.9219	2.0	80	1.9451	0.3063
1.7429	3.0	120	1.8213	0.375
1.5854	4.0	160	1.7126	0.4188
1.4913	5.0	200	1.6547	0.4688
1.3673	6.0	240	1.6200	0.4813
1.2713	7.0	280	1.5822	0.475
1.1907	8.0	320	1.5639	0.4875
1.0516	9.0	360	1.5441	0.4875
1.0037	10.0	400	1.5285	0.4813
0.9538	11.0	440	1.5229	0.4813
0.8983	12.0	480	1.5100	0.4813
0.8616	13.0	520	1.5016	0.4938
0.8417	14.0	560	1.5024	0.4813
0.8078	15.0	600	1.5006	0.4813

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1
• Datasets 2.21.0
• Tokenizers 0.19.1