metadata

license: apache-2.0
base_model: microsoft/beit-large-patch16-224-pt22k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: Psoriasis-500-100aug-224-beit-large
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: validation
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7991266375545851

Psoriasis-500-100aug-224-beit-large

This model is a fine-tuned version of microsoft/beit-large-patch16-224-pt22k on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 1.1823
Accuracy: 0.7991

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
0.8236	0.9973	92	1.1536	0.6358
0.4282	1.9946	184	0.8848	0.7389
0.2305	2.9919	276	0.9811	0.7258
0.1206	4.0	369	0.8858	0.7808
0.1107	4.9973	461	1.1129	0.7397
0.0319	5.9946	553	1.1625	0.7703
0.0073	6.9919	645	1.1938	0.7895
0.0078	8.0	738	1.3031	0.7790
0.0013	8.9973	830	1.2117	0.7974
0.002	9.9729	920	1.1823	0.7991

Classification Report

Class	Precision (%)	Recall (%)	F1-Score (%)	Support
Abnormal	66	62	64	108
Erythrodermic	96	76	85	100
Guttate	95	83	89	114
Inverse	83	91	87	108
Nail	81	84	83	99
Normal	81	79	80	82
Not Define	98	95	96	92
Palm Soles	82	88	85	102
Plaque	70	88	78	84
Psoriatic Arthritis	78	74	76	104
Pustular	71	76	74	112
Scalp	84	86	85	80
Accuracy			82	1185
Macro Avg	82	82	82	1185
Weighted Avg	82	82	82	1185

Framework versions

Transformers 4.41.2
Pytorch 2.1.2
Datasets 2.19.2
Tokenizers 0.19.1