convnextv2-tiny-1k-224-finetuned-crop-neckline

This model is a fine-tuned version of facebook/convnextv2-tiny-1k-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.6160
• Accuracy: 0.8095
• Precision: 0.8101

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: 2e-05
• train_batch_size: 10
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision
No log	1.0	84	1.4226	0.5619	0.5870
No log	2.0	168	1.1924	0.5619	0.5663
No log	3.0	252	0.9542	0.6952	0.7317
No log	4.0	336	0.8255	0.7143	0.7224
No log	5.0	420	0.7614	0.7190	0.7378
1.1937	6.0	504	0.7303	0.7381	0.7454
1.1937	7.0	588	0.6770	0.7667	0.7772
1.1937	8.0	672	0.6849	0.7667	0.7748
1.1937	9.0	756	0.6720	0.7381	0.7532
1.1937	10.0	840	0.7036	0.7286	0.7429
1.1937	11.0	924	0.6752	0.7619	0.7827
0.6846	12.0	1008	0.6399	0.7810	0.7860
0.6846	13.0	1092	0.6860	0.7381	0.7553
0.6846	14.0	1176	0.6827	0.7476	0.7644
0.6846	15.0	1260	0.6160	0.8095	0.8101
0.6846	16.0	1344	0.7032	0.7619	0.7695
0.6846	17.0	1428	0.6916	0.8048	0.8197
0.5051	18.0	1512	0.7070	0.7810	0.7891

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1