{"id":15600623,"url":"https://github.com/borda/kaggle_image-classify","last_synced_at":"2025-08-27T11:08:17.977Z","repository":{"id":40322825,"uuid":"354100280","full_name":"Borda/kaggle_image-classify","owner":"Borda","description":"Various Kaggle image classification challenges solutions","archived":false,"fork":false,"pushed_at":"2025-04-08T11:41:27.000Z","size":13487,"stargazers_count":41,"open_issues_count":2,"forks_count":12,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-05-07T07:19:30.558Z","etag":null,"topics":["colab","dataset","image-classification","kaggle-competition","lightning-flash","notebooks","plant-pathology","pytorch-lightning"],"latest_commit_sha":null,"homepage":"https://borda.github.io/kaggle_plant-pathology","language":"Jupyter Notebook","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/Borda.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":".github/CODEOWNERS","security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2021-04-02T18:22:30.000Z","updated_at":"2025-04-08T11:41:30.000Z","dependencies_parsed_at":"2024-09-18T01:48:21.789Z","dependency_job_id":"47568d12-709d-47a4-bb7b-693fb689aeb1","html_url":"https://github.com/Borda/kaggle_image-classify","commit_stats":{"total_commits":113,"total_committers":7,"mean_commits":"16.142857142857142","dds":"0.34513274336283184","last_synced_commit":"f5e8f854bb26e0766f407e49a1bf6a3b8f67bed5"},"previous_names":[],"tags_count":0,"template":false,"template_full_name":"Borda/kaggle_SandBox","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Borda%2Fkaggle_image-classify","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Borda%2Fkaggle_image-classify/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Borda%2Fkaggle_image-classify/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Borda%2Fkaggle_image-classify/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Borda","download_url":"https://codeload.github.com/Borda/kaggle_image-classify/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":252831323,"owners_count":21810787,"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":["colab","dataset","image-classification","kaggle-competition","lightning-flash","notebooks","plant-pathology","pytorch-lightning"],"created_at":"2024-10-03T02:05:02.730Z","updated_at":"2025-05-07T07:19:54.977Z","avatar_url":"https://github.com/Borda.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Kaggle: Image classification challenges\n\n![CI complete testing](https://github.com/Borda/kaggle_image-classify/workflows/CI%20complete%20testing/badge.svg?branch=main\u0026event=push)\n[![codecov](https://codecov.io/gh/Borda/kaggle_image-classify/branch/main/graph/badge.svg?token=5t1Aj5BIyS)](https://codecov.io/gh/Borda/kaggle_image-classify)\n[![pre-commit.ci status](https://results.pre-commit.ci/badge/github/Borda/kaggle_image-classify/main.svg)](https://results.pre-commit.ci/latest/github/Borda/kaggle_image-classify/main)\n\n## Experimentation\n\n### install this tooling\n\nA simple way how to use this basic functions:\n\n```bash\n! pip install https://github.com/Borda/kaggle_image-classify/archive/main.zip\n```\n\n## Kaggle: [Herbarium 2022](https://www.kaggle.com/competitions/herbarium-2022-fgvc9)\n\nThe Herbarium 2022: Flora of North America dataset comprises 1.05 M images of 15,501 vascular plants, which constitute more than 90% of the taxa documented in North America. The provided dataset is constrained to include only vascular land plants (lycophytes, ferns, gymnosperms, and flowering plants) and it has a long-tail distribution. The number of images per taxon is as few as seven and as many as 100 images. Although more images are available.\n\n![Sample images](./assets/herbarium_sample-imgs.jpg)\n\n### run notebooks in Kaggle\n\n- [🌿Herbarium: EDA 🔎 \u0026 baseline Flash⚡EfficientNet](https://www.kaggle.com/code/jirkaborovec/herbarium-eda-baseline-flash-efficientnet)\n- [🌿Herbarium: Lightning⚡Flash (inference)](https://www.kaggle.com/code/jirkaborovec/herbarium-lightning-flash-inference)\n\n### run notebooks in Colab\n\n- [🌿Herbarium with Lit⚡Flash \u0026 EfficientNet](https://colab.research.google.com/github/Borda/kaggle_image-classify/blob/main/notebooks/Herbarium-with-Flash-EfficientNet.ipynb)\n\n### some results\n\nTraining progress with EffNet-b3 with training for 10 epochs:\n\n![Training process](./assets/herbarium_training-metrics.png)\n\n## Kaggle: [Plant Pathology 2021 - FGVC8](https://www.kaggle.com/c/plant-pathology-2021-fgvc8)\n\nFoliar (leaf) diseases pose a major threat to the overall productivity and quality of apple orchards.\nThe current process for disease diagnosis in apple orchards is based on manual scouting by humans, which is time-consuming and expensive.\n\nThe main objective of the competition is to develop machine learning-based models to accurately classify a given leaf image from the test dataset to a particular disease category, and to identify an individual disease from multiple disease symptoms on a single leaf image.\n\n![Sample images](./assets/plants_sample-images.jpg)\n\n### run notebooks in Kaggle\n\n- [Plant Pathology with Flash](https://www.kaggle.com/jirkaborovec/plant-pathology-with-pytorch-lightning-flash)\n- [Plant Pathology with Lightning ⚡](https://www.kaggle.com/jirkaborovec/plant-pathology-with-lightning)\n- [Plant Pathology with Lightning [predictions]](https://www.kaggle.com/jirkaborovec/plant-pathology-with-lightning-predictions)\n\n### run notebooks in Colab\n\n- [Plant pathology with Lightning](https://colab.research.google.com/github/Borda/kaggle_image-classify/blob/main/notebooks/Plant-Pathology-with-Lightning.ipynb)\n- [Plant pathology with Lightning - StandAlone](https://colab.research.google.com/github/Borda/kaggle_image-classify/blob/main/notebooks/Plant-Pathology-with-Lightning_standalone.ipynb) (without this package)\n- [Plant pathology with Flash](https://colab.research.google.com/github/Borda/kaggle_image-classify/blob/main/notebooks/Plant-Pathology-with-Flash.ipynb)\n\nI would recommend uploading the dataset to you personal gDrive and then in notebooks connect the gDrive which saves you lost of time with re-uploading dataset when ever your Colab is reset... :\\]\n\n### some results\n\nTraining progress with ResNet50 with training for 10 epochs \u003e over 96% validation accuracy:\n\n![Training process](./assets/plants_training-metrics.png)\n\n### More reading\n\n- [Practical Lighting Tips to Rank on Kaggle Image Challenges](https://devblog.pytorchlightning.ai/practical-tips-to-rank-on-kaggle-image-challenges-with-lightning-242e2e533429)\n\n## Kaggle: [iMet Collection 2021 x AIC - FGVC8](https://www.kaggle.com/c/imet-2021-fgvc8)\n\nThe online cataloguing information is generated by subject matter experts and includes a wide range of data. These include, but are not limited to: multiple object classifications, artist, title, period, date, medium, culture, size, provenance, geographic location, and other related museum objects within The Met’s collection.\nAdding fine-grained attributes to aid in the visual understanding of the museum objects will enable the ability to search for visually related objects.\n\n![Sample images](./assets/imet_sample-imgs.png)\n\n### run notebooks in Kaggle\n\n- [iMet Collection with Lightning ⚡](https://www.kaggle.com/jirkaborovec/imet-with-lightning)\n\n### run notebooks in Colab\n\n- [iMet Collection with Lightning with ResNet50](https://colab.research.google.com/github/Borda/kaggle_image-classify/blob/main/notebooks/iMet-with-Lightning.ipynb)\n- [iMet Collection with Lightning and VisionTransformers from TIMM](https://colab.research.google.com/github/Borda/kaggle_image-classify/blob/main/notebooks/iMet-with-Lightning-and-ViT.ipynb)\n\nI would recommend uploading the dataset to you personal gDrive and then in notebooks connect the gDrive which saves you lost of time with re-uploading dataset when ever your Colab is reset... :\\]\n\n### some results\n\nTraining progress with ResNet50 with training for 35 epochs and subset labels with ore then 100 samples:\n\n![training on 100 samples per class](./assets/imet_training-cls-spl-100.png)\n\n## Kaggle: [Cassava Leaf Disease Classification](https://www.kaggle.com/c/cassava-leaf-disease-classification/overview)\n\nThe task is to classify each cassava image into five categories indicating - plant with a certain kind of disease or healthy leaf.\n\nOrganizers introduced a dataset of 21,367 labeled images collected during a regular survey in Uganda. Most images were crowd-sourced from farmers taking photos of their gardens, and annotated by experts at the National Crops Resources Research Institute (NaCRRI) in collaboration with the AI lab at Makerere University, Kampala.\n\n![Sample images](./assets/cassava_images.jpg)\n\n### run notebooks in Colab\n\n- [Cassava with Lightning](https://colab.research.google.com/github/Borda/kaggle_image-classify/blob/main/notebooks/Cassava_with_Lightning.ipynb)\n- [Cassava with Flash](https://colab.research.google.com/github/Borda/kaggle_image-classify/blob/main/notebooks/Cassava_with_Flash.ipynb)\n\nI would recommend uploading the dataset to you personal gDrive and then in notebooks connect the gDrive which saves you lost of time with re-uploading dataset when ever your Colab is reset... :\\]\n\n### some results\n\nTraining progress with ResNet50 with training for 10 epochs:\n\n![Training process](./assets/cassava_metrics.png)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fborda%2Fkaggle_image-classify","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fborda%2Fkaggle_image-classify","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fborda%2Fkaggle_image-classify/lists"}