{"id":22636517,"url":"https://github.com/mrdvince/dermatologist","last_synced_at":"2026-04-16T18:39:27.912Z","repository":{"id":48667898,"uuid":"381366652","full_name":"mrdvince/dermatologist","owner":"mrdvince","description":"Using PyTorch's C++ frontend to visually diagnose melanoma nevus and seborrheic keratosis, using the 2017 ISIC Challenge on Skin Lesion Analysis Towards Melanoma Detection","archived":false,"fork":false,"pushed_at":"2021-07-16T00:37:40.000Z","size":47,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-02-03T14:46:39.370Z","etag":null,"topics":["cpp","deeplearning","other-stuff","pytorch","pytorch-cpp","pytorch-cpp-frontend"],"latest_commit_sha":null,"homepage":"","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/mrdvince.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2021-06-29T13:02:00.000Z","updated_at":"2021-07-16T00:38:34.000Z","dependencies_parsed_at":"2022-09-07T08:41:32.822Z","dependency_job_id":null,"html_url":"https://github.com/mrdvince/dermatologist","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mrdvince%2Fdermatologist","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mrdvince%2Fdermatologist/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mrdvince%2Fdermatologist/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mrdvince%2Fdermatologist/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/mrdvince","download_url":"https://codeload.github.com/mrdvince/dermatologist/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246140547,"owners_count":20729798,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cpp","deeplearning","other-stuff","pytorch","pytorch-cpp","pytorch-cpp-frontend"],"created_at":"2024-12-09T03:22:07.907Z","updated_at":"2026-04-16T18:39:22.883Z","avatar_url":"https://github.com/mrdvince.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n# Skin Cancer Detection\n\nThis is an hobby project to design an algorithm \nthat can visually diagnose 3 classes of skin cancer using PyTorch's C++ frontend.\n\n\u003e Disclaimer: Melanoma is one of the most deadliest forms of skin cancer, so definetly don't use anything on this repo to diagnose yourself.\n\nThese classes include:\n\n- Melanoma \n- Nevus\n- Seborrheic keratosis\n\nThe algorithm will distinguish this malignant skin tumor from two types of benign lesions (nevi and seborrheic keratoses).\n\n\n## Requirements\n\n1. [C++](http://www.cplusplus.com/doc/tutorial/introduction/)\n2. [CMake](https://cmake.org/download/) (minimum version 3.14)\n3. [LibTorch v1.8.0](https://pytorch.org/cppdocs/installing.html)\n4. [Conda](https://docs.conda.io/projects/conda/en/latest/user-guide/install/download.html)\n\n## Getting started\n1. Clone this repo and cd into the cloned directory\n```bash\n  https://github.com/mrdvince/dermatologist.git\n  cd dermatologist\n```\n2. Create a build and cd into it. Then build the project using cmake.\n\nThe build process will download the training, testing and validation datasets (it's a large dataset)\n```bash\nmkdir build \u0026\u0026 cd build\ncmake -DCMAKE_PREFIX_PATH=~/libtorch ..\n```\n3. Build\n```bash\ncmake --build . --config Release\n```\n__Note:__ if the a data folder is not created and the dataset downloaded modify the CMakelists.txt file\nand set the download option to `ON`\n\n4. Run the python convert file included in the cloned folder.\nThis file will download the resnet18 pretrained model and \"trace\" it and save on disc without the final fully connected layer.\n```bash\npython ../convert.py\n```\n5. Finally train the model\n```bash\n./dermatologist resnet18_without_last_layer.pt\n```\n\nThat's pretty much it.\n## Acknowledgements\n\n - [PyTorch C++ examples](https://github.com/pytorch/examples/tree/master/cpp)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmrdvince%2Fdermatologist","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmrdvince%2Fdermatologist","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmrdvince%2Fdermatologist/lists"}