emotion_classification

This model is a fine-tuned version of dennisjooo/emotion_classification on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.7891
• Accuracy: 0.7575

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine_with_restarts
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7123	1.0	25	0.8681	0.735
0.6349	2.0	50	0.8721	0.73
0.6354	3.0	75	0.8732	0.725
0.6189	4.0	100	0.8406	0.735
0.6364	5.0	125	0.8456	0.74
0.5833	6.0	150	0.8503	0.725
0.5384	7.0	175	0.8023	0.755
0.5297	8.0	200	0.8002	0.7525
0.5487	9.0	225	0.8253	0.745
0.5068	10.0	250	0.7891	0.7575

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.1