emotion_classification

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.2445
• Accuracy: 0.5312

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 32
• 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
No log	1.0	10	1.9385	0.325
No log	2.0	20	1.7153	0.4188
No log	3.0	30	1.5905	0.3937
No log	4.0	40	1.4706	0.4625
No log	5.0	50	1.4078	0.5062
No log	6.0	60	1.3739	0.4813
No log	7.0	70	1.3108	0.5125
No log	8.0	80	1.2874	0.5312
No log	9.0	90	1.2810	0.5312
No log	10.0	100	1.2754	0.5437
No log	11.0	110	1.2380	0.5563
No log	12.0	120	1.1721	0.6125
No log	13.0	130	1.2242	0.5875
No log	14.0	140	1.2530	0.525
No log	15.0	150	1.2610	0.575

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3