portrait_cosu_exp3

This model is a fine-tuned version of NekoFi/content-manage-exp2 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.2524
• Accuracy: 0.9149
• Precision: 0.9190
• Recall: 0.9149
• F1: 0.9153
• Confusion Matrix: [[19, 1], [3, 24]]

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: 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: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Confusion Matrix
No log	0.9231	6	0.2921	0.8511	0.8527	0.8511	0.8515	[[17, 3], [4, 23]]
0.5415	2.0	13	0.2564	0.9362	0.9426	0.9362	0.9353	[[17, 3], [0, 27]]
0.5415	2.9231	19	0.3605	0.8723	0.8864	0.8723	0.8730	[[19, 1], [5, 22]]
0.378	3.6923	24	0.2524	0.9149	0.9190	0.9149	0.9153	[[19, 1], [3, 24]]

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1