{"id":48174146,"url":"https://github.com/quangduy201/captionify","last_synced_at":"2026-04-04T17:36:24.239Z","repository":{"id":307253870,"uuid":"797209148","full_name":"quangduy201/captionify","owner":"quangduy201","description":"An AI Image Caption Generator","archived":false,"fork":false,"pushed_at":"2025-08-27T20:40:51.000Z","size":2358,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-08-28T05:15:53.144Z","etag":null,"topics":["cnn","fastapi","huggingface-spaces","image-captioning","kaggle","pytorch","rnn"],"latest_commit_sha":null,"homepage":"https://captionify-app.vercel.app","language":"TypeScript","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/quangduy201.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,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2024-05-07T12:03:46.000Z","updated_at":"2025-08-27T21:03:38.000Z","dependencies_parsed_at":"2025-07-30T09:34:39.562Z","dependency_job_id":null,"html_url":"https://github.com/quangduy201/captionify","commit_stats":null,"previous_names":["quangduy201/captionify"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/quangduy201/captionify","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/quangduy201%2Fcaptionify","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/quangduy201%2Fcaptionify/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/quangduy201%2Fcaptionify/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/quangduy201%2Fcaptionify/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/quangduy201","download_url":"https://codeload.github.com/quangduy201/captionify/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/quangduy201%2Fcaptionify/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31407645,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-04T10:20:44.708Z","status":"ssl_error","status_checked_at":"2026-04-04T10:20:06.846Z","response_time":60,"last_error":"SSL_read: 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":["cnn","fastapi","huggingface-spaces","image-captioning","kaggle","pytorch","rnn"],"created_at":"2026-04-04T17:36:24.129Z","updated_at":"2026-04-04T17:36:24.211Z","avatar_url":"https://github.com/quangduy201.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Captionify - An AI Image Caption Generator\n\n**Captionify** is a simple web application that generates English captions from images using a deep learning model.\nIt combines a **FastAPI backend** (for model inference) and a responsive **HTML/CSS/JS frontend**.\n\n\u003e The model is trained on datasets like **Flickr8k** with a **CNN + RNN architecture**.\n\n## Features\n\n- Upload images drag \u0026 drop or file browser.\n- Generate descriptive English captions from your images.\n- Live typing effect for generated captions.\n- Trained on Flickr8k dataset with a custom PyTorch model.\n- API endpoint `POST /reload-model` to dynamically reload the latest model from Kaggle.\n\n\n## Setup\n\n### 1. Clone the repository:\n```shell\ngit clone https://github.com/quangduy201/captionify.git\ncd captionify\n```\n\n### 2. Create and activate a virtual environment:\n```shell\n# Create a virtual environment named '.venv'\npython -m venv .venv\n\n# Activate the virtual environment\n# On Windows\n.venv\\Scripts\\activate\n# On macOS/Linux\nsource .venv/bin/activate\n```\n   \n### 3. Install dependencies:\n```shell\npip install -r requirements.txt\n```\n    \n### 4. Run:\n```shell\nuvicorn run:app --reload\n```\n\nThis will:\n- Automatically download the latest model from [`Kaggle Model Hub`](https://www.kaggle.com/models/quangduy201/image-captioning).\n- Load model + vocabulary into memory.\n- Start FastAPI on [`http://localhost:8000`](http://localhost:8000)\n\n\n### 5. Access the application:\nOpen a web browser and go to [`http://localhost:8000`](http://localhost:8000) to access the application.\n\n\n## Train your own model\n\nYou can train a custom captioning model using the provided Kaggle notebook [here](https://www.kaggle.com/code/quangduy201/image-captioning-pytorch)\n### 1. Open the notebook:\nVisit the provided Kaggle link and create a copy of the notebook.\n\n### 2. Choose your suitable dataset:\nYou can choose which dataset which is the most suitable for your model.\nThe default dataset is [flickr8k](https://www.kaggle.com/datasets/quangduy201/flickr8k).\n\n### 2. Settings for the notebook's session\nYou should train the model using GPU T4 x2 or GPU P100\n\n### 3. Run the notebook:\nSimply press `Run all` in the notebook to train your custom model.\n\n### 4. Download the trained model:\nAfter training, download the trained model checkpoint (`/kaggle/working/training/output/checkpoint.pth.tar`).\n\n### 5. Place the trained model:\nPlace the downloaded `checkpoint.pth.tar` file in the `training/output` directory of the repository.\n\n### 6. Improve the trained model (Optional):\nIf you want to improve the trained model, you can upload your current checkpoint of the model\nand use it as an Input of the notebook.\n\n\n## Dependencies\n\n- fastapi\n- uvicorn\n- torch\n- torchvision\n- spacy\n- tqdm\n- Pillow\n- python-multipart\n- tensorboard\n- kagglehub\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fquangduy201%2Fcaptionify","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fquangduy201%2Fcaptionify","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fquangduy201%2Fcaptionify/lists"}