{"id":13936081,"url":"https://github.com/svishnu88/TGS-SaltNet","last_synced_at":"2025-07-19T21:31:37.660Z","repository":{"id":152204662,"uuid":"153910613","full_name":"svishnu88/TGS-SaltNet","owner":"svishnu88","description":"Kaggle | 21st place solution for TGS Salt Identification Challenge","archived":false,"fork":false,"pushed_at":"2019-03-11T06:09:22.000Z","size":28,"stargazers_count":83,"open_issues_count":1,"forks_count":24,"subscribers_count":7,"default_branch":"master","last_synced_at":"2024-11-27T04:30:33.864Z","etag":null,"topics":["computer-vision","deep-learning","fastai","kaggle-competition","pytorch"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/svishnu88.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2018-10-20T13:52:50.000Z","updated_at":"2024-03-31T15:09:12.000Z","dependencies_parsed_at":null,"dependency_job_id":"7bda174d-3734-4e9b-810e-fcd4b25b874d","html_url":"https://github.com/svishnu88/TGS-SaltNet","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/svishnu88/TGS-SaltNet","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/svishnu88%2FTGS-SaltNet","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/svishnu88%2FTGS-SaltNet/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/svishnu88%2FTGS-SaltNet/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/svishnu88%2FTGS-SaltNet/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/svishnu88","download_url":"https://codeload.github.com/svishnu88/TGS-SaltNet/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/svishnu88%2FTGS-SaltNet/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":266019657,"owners_count":23864916,"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":["computer-vision","deep-learning","fastai","kaggle-competition","pytorch"],"created_at":"2024-08-07T23:02:21.906Z","updated_at":"2025-07-19T21:31:32.647Z","avatar_url":"https://github.com/svishnu88.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# TGS-SaltNet\nKaggle | 21st place solution for TGS Salt Identification Challenge\n\n## General\n\nI recently participated in a Kaggle competition [TGS Salt Identification Challenge](https://www.kaggle.com/c/tgs-salt-identification-challenge)\nand reached the 21st place. This repository contains the final code which resulted in the best model. The code demonstrates usage of different important techniques using [fast.ai](http://www.fast.ai/) and [PyTorch](https://pytorch.org/).\n1. Use ResNet model as an encoder for UNet. \n2. Add intermediate layers like [BAM](http://bmvc2018.org/contents/papers/0092.pdf),[Squeeze \u0026 Excitation](https://arxiv.org/abs/1803.02579) blocks in a ResNet34 model which can be easily replicated for other network architectures.\n3. Show how to add [Deep supervision](https://www.kaggle.com/c/tgs-salt-identification-challenge/discussion/65933) to the network, and calculate loss and combine loss at different scale. \n\n## Main software used\n\n1. fastai - 0.7\n2. pytorch - 0.4\n3. python - 3.6\n\n## Hardware required\n\nThe code was tested with TitanX GPU/1080ti.\n\n## Thanks\n\nA special thanks to Heng for his generous contributions to different ideas in the competition, for a long list of amazing Kaglle community members, Jeremy and Fast.ai community for the amazing and flexible fastai framework. \n\n \n\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsvishnu88%2FTGS-SaltNet","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsvishnu88%2FTGS-SaltNet","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsvishnu88%2FTGS-SaltNet/lists"}