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: 1.9455
• Accuracy: 0.3

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: 1e-05
• 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
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	10	2.0636	0.1187
No log	2.0	20	2.0568	0.1437
No log	3.0	30	2.0321	0.1812
No log	4.0	40	2.0247	0.2
No log	5.0	50	1.9975	0.3125
No log	6.0	60	1.9793	0.2875
No log	7.0	70	1.9746	0.275
No log	8.0	80	1.9530	0.3063
No log	9.0	90	1.9487	0.3438
No log	10.0	100	1.9513	0.2812

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3