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: 1.6256
Accuracy: 0.5625

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: 0.00025
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	10	1.7794	0.4875
No log	2.0	20	1.6813	0.4938
0.2276	3.0	30	1.7602	0.4875
0.2276	4.0	40	1.9172	0.4562
0.2048	5.0	50	1.9316	0.4625
0.2048	6.0	60	1.8285	0.5
0.2048	7.0	70	1.6341	0.5687
0.1617	8.0	80	1.7461	0.5375
0.1617	9.0	90	1.6544	0.5312
0.1766	10.0	100	1.9449	0.4875
0.1766	11.0	110	1.7565	0.5125
0.1766	12.0	120	1.8936	0.5
0.1979	13.0	130	1.6812	0.5687
0.1979	14.0	140	1.7619	0.5188
0.1694	15.0	150	1.6903	0.55

Framework versions

Transformers 4.33.1
Pytorch 1.12.1+cu116
Datasets 2.4.0
Tokenizers 0.12.1

dini-r-a/emotion_classification

emotion_classification

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

dini-r-a/
emotion_classification