Seven lesion categories
According to the model card, the model classifies benign keratosis-like lesions, basal cell carcinoma, actinic keratoses, vascular lesions, melanocytic nevi, melanoma, and dermatofibroma.
Open Source Model Profile · Anwarkh1
Skin_Cancer-Image_Classification
Skin Cancer Image Classification is an 85.8M-parameter ViT image classifier from Anwarkh1. According to the model card, it distinguishes seven skin-lesion categories and reports 0.9695 final validation accuracy.
- Publisher
- Anwarkh1
- Task
- image-classification
- Model type
- vit
- License
- apache-2.0
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
Skin Cancer Image Classification is published by Anwarkh1 as an image-classification fine-tune. The captured configuration identifies ViTForImageClassification with a vit model type, and Safetensors metadata reports 85804039 parameters. According to the model card, it adapts Google's 16x16-patch ViT pretrained on ImageNet21k with a replaced classification head for skin-cancer images.
Recorded capabilities
ViT ImageNet21k starting point
The publisher describes Google's ViT with 16x16 patch size trained on ImageNet21k, adapted with a replaced classification head.
Published epoch metrics
According to the model card, epoch five reports 0.1208 train loss, 0.9614 train accuracy, 0.1000 validation loss, and 0.9695 validation accuracy.
Documented Skin Cancer dataset
The card names Marmal88's Skin Cancer Dataset on Hugging Face as the source.
Use cases in the source record
- Skin-lesion image classification research across the seven publisher-listed categories, including melanoma, basal cell carcinoma, and melanocytic nevi.
- Fine-tuning or evaluation experiments that reuse the documented Adam, cross-entropy, batch-32, five-epoch procedure.
Limitations and unknowns
- Training information comes from the publisher model card and has not been independently verified by Ethen.
- This is a research classification record, not validated clinical software; no clinical evaluation is provided in the current evidence.
- Provider state is historical snapshot data, not independently refreshed current availability.
Source and provenance
Source: Anwarkh1/Skin_Cancer-Image_Classification
Captured: Unknown. Processed: 2026-09-07T19:34:28.954128+00:00.
Skin Cancer Image Classification Model Introduction This model is designed for the classification of skin cancer images into various categories including benign keratosis-like lesions, basal cell carcinoma, actinic keratoses, vascular lesions, melanocytic nevi, melanoma, and dermatofibroma. Model Overview Model Architecture: Vision Transformer (ViT) Pre-trained Model: Google's ViT with 16x16 patch size and trained on ImageNet21k dataset Modified Classification Head: The classification head has been replaced to adapt the model to the skin cancer classification task. Dataset Dataset Name: Skin Cancer Dataset Source: Marmal88's Skin Ca…
F001F002F003F004F005F006F007F008F010F011F012F013F014F015