{"id":16933038,"url":"https://github.com/zlsh80826/image-caption-tf","last_synced_at":"2025-04-11T18:41:41.742Z","repository":{"id":39739921,"uuid":"160311458","full_name":"zlsh80826/image-caption-tf","owner":"zlsh80826","description":"Image Caption","archived":false,"fork":false,"pushed_at":"2023-03-24T22:20:20.000Z","size":17927,"stargazers_count":17,"open_issues_count":2,"forks_count":1,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-03-25T14:39:02.905Z","etag":null,"topics":["cs565600","image-caption","image-captioning","kaggle-competition","tensorflow"],"latest_commit_sha":null,"homepage":null,"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/zlsh80826.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":"2018-12-04T06:42:38.000Z","updated_at":"2022-10-21T14:28:34.000Z","dependencies_parsed_at":"2025-02-19T23:43:12.093Z","dependency_job_id":null,"html_url":"https://github.com/zlsh80826/image-caption-tf","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/zlsh80826%2Fimage-caption-tf","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zlsh80826%2Fimage-caption-tf/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zlsh80826%2Fimage-caption-tf/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zlsh80826%2Fimage-caption-tf/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/zlsh80826","download_url":"https://codeload.github.com/zlsh80826/image-caption-tf/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248460754,"owners_count":21107547,"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":["cs565600","image-caption","image-captioning","kaggle-competition","tensorflow"],"created_at":"2024-10-13T20:48:22.736Z","updated_at":"2025-04-11T18:41:41.705Z","avatar_url":"https://github.com/zlsh80826.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Image Caption Tensorflow\n\n* Image caption model base on [\nShow and Tell: A Neural Image Caption Generator](https://arxiv.org/abs/1411.4555) with some modifications.\n* The dataset come from [Microsoft COCO 2014](http://cocodataset.org/#home) train and valid, and we do some redistribution.\n* This model is trained for [NTHU CS565600](http://www.cs.nthu.edu.tw/~shwu/) image caption competition.\n* Our model achieved 0.944 CIDEr-D score on single model, which is the 1st place of the [Image Caption Kaggle Competition](https://www.kaggle.com/c/datalabcup-image-caption-2018fall/leaderboard).\n* We provide end to end scripts and pretrained weight for reproduction.\n* [This slides](https://docs.google.com/presentation/d/1LU6CEDQIag6S8wtL4aTgqb9HuYRfsg2jOsGEOAhkN5g/present?slide=id.p) briefly describe the implementation\n* If you meet any problem, feel free to contact (zlsh80826@gmail.com).\n\n## Requirements\n\nHere are some required libraries.\n\n### General \n* python \u003e= 3.6\n* cuda \u003e= 10.0 (or base on your tensorflow version)\n\n### Python\n* please refer requirements.txt\n\n## Reproduce from scratch\n\n### Download the data\n\n```Bash\ncd data\nsh download.sh\n```\n\n### Redistribute the data (Competition required)\n\n```Bash\npython split.py\n```\n\n### Generate the image features\n\nWe use the [NASNet](https://arxiv.org/abs/1707.07012) model [pretrained by Keras](https://keras.io/applications/) \nto get the image features. This step may took over one hour.\n\n```Bash\npython nasnet.py\n```\n\n### Create tensorflow records\n\n```Bash\ncd ../script\npython create_tfrecord.py\n```\n\n### Train\n\n```Bash\npython train.py\n```\n\n### Evaluate on validation set\n\n```Bash\npython inference.py\n```\n\n## Performance\n\n||CIDEr-D|\n|---|---|\n|Single Model|0.944|\n|Ensemble Model|0.955|\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzlsh80826%2Fimage-caption-tf","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fzlsh80826%2Fimage-caption-tf","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzlsh80826%2Fimage-caption-tf/lists"}