Train-Augmentation-Psoriasis-Project

This model is a fine-tuned version of ahmedesmail16/Psoriasis-Project-Aug-M2-swinv2-base-patch4-window12-192-22k on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0793
• Accuracy: 0.8182

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4867	0.99	93	0.8388	0.7075
0.2164	2.0	187	0.8105	0.7470
0.1503	2.99	280	0.8719	0.7470
0.1074	4.0	374	0.8725	0.7708
0.0649	4.99	467	0.8323	0.7984
0.0259	6.0	561	1.2222	0.7945
0.0162	6.99	654	1.0419	0.8024
0.0135	8.0	748	0.9799	0.8300
0.0031	8.99	841	1.1125	0.8063
0.0041	9.95	930	1.0793	0.8182

Test results

class	precision	recall	f1-score	support
Erythrodermic	0.75	0.86	0.80	7
Guttate	0.91	0.87	0.89	23
Inverse	1.00	0.82	0.90	17
Nail	0.95	0.95	0.95	20
Normal	0.95	0.80	0.87	25
Not Define	1.00	1.00	1.00	29
Palm Soles	0.83	0.90	0.86	21
Plaque	0.90	0.76	0.83	25
Psoriatic Arthritis	0.73	1.00	0.84	8
Pustular	0.88	0.82	0.85	17
Scalp	0.87	0.83	0.85	24
UPNormal	0.82	0.94	0.87	52
accuracy			0.88	268
macro avg	0.88	0.88	0.88	268
weighted avg	0.89	0.88	0.88	268

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.19.1
• Tokenizers 0.15.2