ConvNext-V2-Retinopathy

This model is a fine-tuned version of syedmuhammad/ConvNextV2-Diabetec-Retinopathy on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.0219
• Accuracy: 0.9901

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.125	1.0	113	0.0339	0.9901
0.2206	2.0	227	0.0139	0.9901
0.1751	3.0	340	0.0114	0.9950
0.0599	4.0	454	0.0277	0.9950
0.1122	5.0	567	0.0328	0.9950
0.093	6.0	681	0.0240	0.9901
0.0673	7.0	794	0.0251	0.9950
0.0718	8.0	908	0.0458	0.9851
0.0632	9.0	1021	0.0477	0.9901
0.0263	10.0	1135	0.0399	0.9950
0.0304	11.0	1248	0.0295	0.9901
0.0892	12.0	1362	0.0330	0.9950
0.0227	13.0	1475	0.0287	0.9901
0.0253	14.0	1589	0.0262	0.9901
0.1242	14.93	1695	0.0219	0.9901

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1