convnextv2-tiny-1k-224-finetuned-hand-final

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

• Loss: 0.6638
• Accuracy: 0.7563

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: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6669	1.0	14	0.6050	0.6834
0.5796	2.0	28	0.5599	0.7362
0.5417	3.0	42	0.5486	0.7437
0.5466	4.0	56	0.5528	0.7387
0.5213	5.0	70	0.5673	0.7462
0.493	6.0	84	0.5432	0.7613
0.5051	7.0	98	0.5457	0.7513
0.4656	8.0	112	0.5444	0.7563
0.4399	9.0	126	0.5430	0.7613
0.4213	10.0	140	0.5507	0.7613
0.4118	11.0	154	0.5619	0.7538
0.4015	12.0	168	0.5383	0.7513
0.3785	13.0	182	0.5567	0.7563
0.3487	14.0	196	0.5972	0.7462
0.3401	15.0	210	0.6059	0.7462
0.3215	16.0	224	0.6051	0.7563
0.3171	17.0	238	0.6228	0.7513
0.2971	18.0	252	0.6529	0.7563
0.3111	19.0	266	0.6309	0.7588
0.2722	20.0	280	0.6444	0.7588
0.2677	21.0	294	0.6373	0.7588
0.2721	22.0	308	0.6393	0.7538
0.2694	23.0	322	0.6382	0.7613
0.2731	24.0	336	0.6543	0.7613
0.257	25.0	350	0.6638	0.7563

Framework versions

• Transformers 4.33.0
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3