{"id":18811303,"url":"https://github.com/activeloopai/omdena-aerial","last_synced_at":"2025-07-13T17:09:09.891Z","repository":{"id":87355465,"uuid":"294724139","full_name":"activeloopai/omdena-aerial","owner":"activeloopai","description":null,"archived":false,"fork":false,"pushed_at":"2020-09-11T15:36:56.000Z","size":811,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-07-10T14:55:13.095Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","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/activeloopai.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":"2020-09-11T14:55:35.000Z","updated_at":"2021-02-28T12:06:00.000Z","dependencies_parsed_at":null,"dependency_job_id":"1cc7d8d9-de7c-45ba-9a83-c086649b3d4a","html_url":"https://github.com/activeloopai/omdena-aerial","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/activeloopai/omdena-aerial","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/activeloopai%2Fomdena-aerial","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/activeloopai%2Fomdena-aerial/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/activeloopai%2Fomdena-aerial/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/activeloopai%2Fomdena-aerial/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/activeloopai","download_url":"https://codeload.github.com/activeloopai/omdena-aerial/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/activeloopai%2Fomdena-aerial/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":265175567,"owners_count":23722661,"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":[],"created_at":"2024-11-07T23:25:34.655Z","updated_at":"2025-07-13T17:09:09.863Z","avatar_url":"https://github.com/activeloopai.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Omdena Aerial\n\nBased on :- https://github.com/jmather625/predicting-poverty-replication\n\n\nThe above repository involves a lot of manual work to get the data. Using hub, we can eliminate all these tedious tasks.\nHub also allows you to stream data insteading of loading all of it your RAM, thus allowing you to train models on large datasets.\nIt's ideal for distributed teams that work on shared datasets.\n\nThe entire dataset is now stored on hub and can be easily accessed with a couple of lines of code.\nYou can also view the entire dataset at https://app.activeloop.ai/dataset/omdena/predicting-poverty-replication-full.\n\nThe original repository consisted of 5 notebooks, the first two of which can now be skipped and we can jump directly to the cnn training notebook.\nThe original notebooks, train_cnn.ipynb and feature_extract.ipynb have been slightly modified to load data from hub.\nRunning the entire repository is now as simple as running the following 3 notebooks in order. \n```\ntrain_cnn_hub.ipynb\nfeature_extract_hub.ipynb\npredict_consumption.ipynb\n```\n**P.S. Ensure that the data folder is present in the base directory.**\n\n\n\nUnderstanding how the data was stored is important in case you want to adapt this approach for future models or make modifications to the current one. Go through \"store_omdena.py\" for the same. \n\nIf you want to try out the storage code:-\n```\nInstall hub, register and login \n\u003e pip3 install hub\n\u003e hub register\n\u003e hub login\nCreate a new folder in data folder named \"countries\". \nIn the countries folder create 3 folders, namely ethiopia_2015, malawi_2016 and nigeria_2015.\nCreate a subfolder images in each of the three folders. \nPlace the images that need to be stored in the images folder of their corresponding country.\n\u003e python3 store_omdena.py\"\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Factiveloopai%2Fomdena-aerial","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Factiveloopai%2Fomdena-aerial","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Factiveloopai%2Fomdena-aerial/lists"}