{"id":40676251,"url":"https://github.com/alcunha/geo_prior_tf","last_synced_at":"2026-01-21T09:40:40.985Z","repository":{"id":169437335,"uuid":"360697345","full_name":"alcunha/geo_prior_tf","owner":"alcunha","description":"This is an unofficial TensorFlow implementation of Presence-Only Geographical Priors for Fine-Grained Image Classification","archived":false,"fork":false,"pushed_at":"2022-10-31T19:11:50.000Z","size":88,"stargazers_count":0,"open_issues_count":0,"forks_count":1,"subscribers_count":2,"default_branch":"master","last_synced_at":"2024-01-27T20:07:10.122Z","etag":null,"topics":["deep-learning","geo-prior","inaturalist","tensorflow"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/alcunha.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,"governance":null}},"created_at":"2021-04-22T22:26:22.000Z","updated_at":"2022-10-31T19:11:55.000Z","dependencies_parsed_at":"2023-10-29T13:33:41.742Z","dependency_job_id":null,"html_url":"https://github.com/alcunha/geo_prior_tf","commit_stats":{"total_commits":36,"total_committers":1,"mean_commits":36.0,"dds":0.0,"last_synced_commit":"0e5d2744f63a2aa3a6a6b48c7a8f15ff189f8073"},"previous_names":["alcunha/geo_prior_tf"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/alcunha/geo_prior_tf","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alcunha%2Fgeo_prior_tf","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alcunha%2Fgeo_prior_tf/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alcunha%2Fgeo_prior_tf/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alcunha%2Fgeo_prior_tf/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/alcunha","download_url":"https://codeload.github.com/alcunha/geo_prior_tf/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alcunha%2Fgeo_prior_tf/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28631544,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-21T04:47:28.174Z","status":"ssl_error","status_checked_at":"2026-01-21T04:47:22.943Z","response_time":86,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"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":["deep-learning","geo-prior","inaturalist","tensorflow"],"created_at":"2026-01-21T09:40:40.254Z","updated_at":"2026-01-21T09:40:40.978Z","avatar_url":"https://github.com/alcunha.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"This is an unofficial TensorFlow implementation of [Presence-Only Geographical Priors for Fine-Grained Image Classification](https://arxiv.org/abs/1906.05272)\n\n### Requirements\n\nPrepare an environment with python=3.8, tensorflow=2.3.1.\n\nDependencies can be installed using the following command:\n```bash\npip install -r requirements.txt\n```\n\n### Data\n\nPlease refer to the [iNat 2018 Github page](https://github.com/visipedia/inat_comp/tree/master/2018) for additional dataset details and download links.\n\nThe original CNN predictions file used for evaluation can be downloaded from the official project [website](http://www.vision.caltech.edu/~macaodha/projects/geopriors/index.html).\n\n### Training\n\nTo train a geo prior model use the script `train.py`:\n```bash\npython train.py --train_data_json=PATH_TO_BE_CONFIGURED/train2018.json \\\n    --train_location_info_json=PATH_TO_BE_CONFIGURED/train2018_locations.json \\\n    --val_data_json=PATH_TO_BE_CONFIGURED/val2018.json \\\n    --val_location_info_json=PATH_TO_BE_CONFIGURED/val2018_locations.json \\\n    --model_dir=PATH_TO_BE_CONFIGURED/geo_prior_ckp/ \\\n    --random_seed=42\n```\n\nOther training hyperparams can also be passed as flags. For more parameter information, please refer to `train.py`.\n\n### Evaluation\n\nTo evaluate a model use the script `eval.py`:\n```bash\npython eval.py --test_data_json=PATH_TO_BE_CONFIGURED/val2018.json \\\n    --test_location_info_json=PATH_TO_BE_CONFIGURED/val2018_locations.json \\\n    --cnn_predictions_file=PATH_TO_BE_CONFIGURED/inat2018_val_preds_sparse.npz \\\n    --ckpt_dir=PATH_TO_BE_CONFIGURED/geo_prior_ckp/\n```\n\n### Results\n\n| Prior                                   | Classifier* | Dataset  | Accuracy |\n|-----------------------------------------|-------------|----------|----------|\n| No Prior [1]                            | InceptionV3 | iNat2018 | 60.20    |\n| Geo Prior (no photographer) [1]         | InceptionV3 | iNat2018 | 72.84    |\n| Geo Prior (no photographer) [(ours)](https://drive.google.com/file/d/1get9lJEK2jw3qKPIXNd6037Kwz0kQQbC/view?usp=sharing)  | InceptionV3 | iNat2018 | 72.94    |\n| Geo Prior (full) [1]                    | InceptionV3 | iNat2018 | 72.68    |\n| Geo Prior (full) [(ours)](https://drive.google.com/file/d/1I5tSBM1LxZgGLJZFGq-IILepKQocnL1d/view?usp=sharing)             | InceptionV3 | iNat2018 | 72.84    |\n\n*Classifier predictions are from the original paper [1].\n\n### Source\n\n[1] Original paper: https://arxiv.org/abs/1906.05272\n\n[2] Official PyTorch code: https://github.com/macaodha/geo_prior\n\n### Contact\n\nIf you have any questions, feel free to contact Fagner Cunha (e-mail: fagner.cunha@icomp.ufam.edu.br) or Github issues. \n\n### License\n\n[Apache License 2.0](LICENSE)","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Falcunha%2Fgeo_prior_tf","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Falcunha%2Fgeo_prior_tf","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Falcunha%2Fgeo_prior_tf/lists"}