convnextv2-large-1k-224-finetuned-Lesion-Classification-HAM10000-AH-60-20-20-Shuffled-3rd

This model is a fine-tuned version of facebook/convnextv2-large-1k-224 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0672
• Accuracy: 0.9893

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: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.9
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.9123	1.0	114	1.9145	0.2241
1.7059	2.0	229	1.6895	0.5599
1.3919	3.0	343	1.3194	0.6634
1.0811	4.0	458	1.0303	0.7151
0.9301	5.0	572	0.8376	0.7562
0.7336	6.0	687	0.6449	0.8374
0.5805	7.0	801	0.5091	0.8736
0.5982	8.0	916	0.4677	0.8826
0.4939	9.0	1030	0.4029	0.8842
0.3296	10.0	1145	0.3051	0.9204
0.3559	11.0	1259	0.5266	0.8366
0.3089	12.0	1374	0.3569	0.8883
0.2272	13.0	1488	0.1487	0.9680
0.3355	14.0	1603	0.2344	0.9269
0.1581	14.93	1710	0.0672	0.9893

Framework versions

• Transformers 4.32.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3