vit-base-patch16-224-finetuned-Visual-Emotional

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

• Loss: 1.3141
• 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: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 32

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.87	5	2.1419	0.15
2.1722	1.91	11	2.0381	0.1625
2.1722	2.96	17	1.8729	0.3
1.8696	4.0	23	1.6683	0.3625
1.8696	4.87	28	1.5172	0.4
1.4531	5.91	34	1.3960	0.4625
1.1483	6.96	40	1.3788	0.45
1.1483	8.0	46	1.3186	0.5125
0.955	8.87	51	1.2741	0.475
0.955	9.91	57	1.2990	0.5
0.7894	10.96	63	1.2462	0.475
0.7894	12.0	69	1.3090	0.5375
0.6769	12.87	74	1.2809	0.5125
0.5958	13.91	80	1.3020	0.525
0.5958	14.96	86	1.3032	0.5
0.5179	16.0	92	1.2624	0.5375
0.5179	16.87	97	1.2776	0.525
0.4808	17.91	103	1.2705	0.525
0.4808	18.96	109	1.2792	0.5125
0.4025	20.0	115	1.2923	0.5375
0.3908	20.87	120	1.3156	0.525
0.3908	21.91	126	1.3290	0.5375
0.3384	22.96	132	1.3141	0.5625
0.3384	24.0	138	1.3253	0.55
0.3428	24.87	143	1.3502	0.5375
0.3428	25.91	149	1.3498	0.525
0.3236	26.96	155	1.3450	0.525
0.2951	27.83	160	1.3425	0.5375

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0