thenewsupercell/me_Emotion_DF_Image_VIT_V2

Louis_Emotion_DF_Image_VIT_V2

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: 0.9591
• Accuracy: 0.6776

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-06
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.148	1.0	1795	1.1148	0.5982
0.8387	2.0	3590	1.0005	0.6411
0.8989	3.0	5385	0.9685	0.6486
0.8048	4.0	7180	0.9354	0.6559
0.6871	5.0	8975	0.9494	0.6559
0.5955	6.0	10770	0.9346	0.6693
0.472	7.0	12565	0.9493	0.6746
0.4086	8.0	14360	0.9603	0.6776
0.3915	9.0	16155	0.9929	0.6773
0.3441	10.0	17950	1.0307	0.6760
0.3019	11.0	19745	1.0561	0.6768
0.3528	12.0	21540	1.0845	0.6743
0.1964	13.0	23335	1.1124	0.6734
0.3125	14.0	25130	1.1289	0.6734
0.1854	15.0	26925	1.1372	0.6704

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0