README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: Output-prova_melanoma
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9466666666666667

Output-prova_melanoma

This model is a fine-tuned version of UnipaPolitoUnimore/vit-large-patch32-384-melanoma on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.2202
• Accuracy: 0.9467

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: 1.5e-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: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4501	1.0	47	0.2094	0.9667
0.5554	2.0	94	0.2010	0.9733
0.5299	3.0	141	0.1595	0.9733
0.0854	4.0	188	0.1529	0.9667
0.2766	5.0	235	0.1466	0.9667
0.3158	6.0	282	0.1916	0.96
0.1322	7.0	329	0.1924	0.9733
0.065	8.0	376	0.1905	0.9533
0.1565	9.0	423	0.2025	0.9467
0.1296	10.0	470	0.2367	0.9333
0.2448	11.0	517	0.2255	0.94
0.067	12.0	564	0.2315	0.94
0.0764	13.0	611	0.2479	0.9467
0.1472	14.0	658	0.2599	0.9333
0.0483	15.0	705	0.1911	0.9533
0.0961	16.0	752	0.1869	0.9533
0.1146	17.0	799	0.2355	0.9333
0.2117	18.0	846	0.1930	0.94
0.2859	19.0	893	0.1902	0.9467
0.0798	20.0	940	0.2436	0.9333
0.16	21.0	987	0.2341	0.94
0.1968	22.0	1034	0.3552	0.9067
0.1049	23.0	1081	0.2541	0.9267
0.1102	24.0	1128	0.1839	0.9467
0.3039	25.0	1175	0.2269	0.9333
0.1188	26.0	1222	0.2063	0.9533
0.2008	27.0	1269	0.1972	0.94
0.1113	28.0	1316	0.2157	0.94
0.1377	29.0	1363	0.2031	0.9533
0.042	30.0	1410	0.2124	0.9533
0.0841	31.0	1457	0.2174	0.94
0.046	32.0	1504	0.2136	0.9467
0.1309	33.0	1551	0.1981	0.96
0.1207	34.0	1598	0.2334	0.94
0.1216	35.0	1645	0.2238	0.94
0.0518	36.0	1692	0.2441	0.9467
0.0852	37.0	1739	0.2243	0.9467
0.0853	38.0	1786	0.2028	0.9533
0.055	39.0	1833	0.2124	0.9467
0.0646	40.0	1880	0.2202	0.9467

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 1.13.1+cu116
• Datasets 2.11.0
• Tokenizers 0.13.3