|
|
---
|
|
|
license: mit
|
|
|
---
|
|
|
|
|
|
# ArtifyAI - COCO Dataset Downloader and Organizer |
|
|
|
|
|
Welcome to ArtifyAI, a project designed to download images from the COCO dataset and organize them into your Google Drive for further use in machine learning and computer vision projects. |
|
|
|
|
|
## Introduction |
|
|
|
|
|
This project automates the process of downloading a subset of images from the COCO dataset and storing them directly in Google Drive. It's aimed at researchers and developers who need a quick way to set up their dataset for training image-based models. The notebook is structured to be easy to follow, even for non-technical users. |
|
|
|
|
|
## Requirements |
|
|
|
|
|
To use this notebook, you need: |
|
|
|
|
|
- Google Colab (optional, but recommended for ease of use) |
|
|
- Google Drive (to store the dataset) |
|
|
- Basic knowledge of Python (optional) |
|
|
- Internet connection (for downloading images) |
|
|
- Libraries: |
|
|
- `pycocotools` |
|
|
- `requests` |
|
|
- `tqdm` |
|
|
- `matplotlib` |
|
|
- `numpy` |
|
|
- `shutil` |
|
|
|
|
|
## Installation |
|
|
|
|
|
1. **Clone the repository (optional)**: |
|
|
```bash |
|
|
git clone https://github.com/your-repo/ArtifyAI.git |
|
|
cd ArtifyAI |
|
|
``` |
|
|
|
|
|
2. **Open the notebook**: |
|
|
- You can directly upload the notebook (`ArtifyAI_v1_0.ipynb`) to your Google Colab environment or use it locally in your Python environment. |
|
|
|
|
|
3. **Install dependencies**: |
|
|
If you're running this locally, make sure to install the required libraries: |
|
|
```bash |
|
|
pip install pycocotools tqdm matplotlib numpy requests |
|
|
``` |
|
|
|
|
|
## Running the Project |
|
|
|
|
|
Once all dependencies are installed, follow these steps: |
|
|
|
|
|
1. **Open Google Colab**: |
|
|
- If you're using Google Colab, upload the notebook to your Colab environment. |
|
|
|
|
|
2. **Mount Google Drive**: |
|
|
- The notebook will prompt you to mount your Google Drive. Ensure your account is connected so that the images are downloaded and saved to a designated folder in your drive. |
|
|
|
|
|
3. **Run the Code**: |
|
|
- Execute the cells sequentially to start downloading the COCO dataset images. The notebook uses `pycocotools` to fetch image URLs from the COCO API and downloads them using `requests`. Progress is tracked with `tqdm`. |
|
|
|
|
|
4. **Transfer Images to Google Drive**: |
|
|
- After downloading, the images will automatically be moved to a folder in your Google Drive (`MyDrive/coco_dataset`). |
|
|
|
|
|
## Using Google Colab |
|
|
|
|
|
For non-technical users, we recommend using [Google Colab](https://colab.research.google.com/). It provides a cloud-based environment where you can run the notebook without installing Python or any dependencies on your local machine. |
|
|
|
|
|
1. **Upload the Notebook**: |
|
|
- Simply drag and drop the `.ipynb` file into Colab. |
|
|
|
|
|
2. **Mount Google Drive**: |
|
|
- The notebook includes a step to mount Google Drive for file storage. Follow the on-screen instructions to authorize access. |
|
|
|
|
|
3. **Run the Notebook**: |
|
|
- Execute each cell in the notebook by clicking the play button next to each cell. Ensure all code cells are run in order. |
|
|
|
|
|
## Moving Files to Google Drive |
|
|
|
|
|
After downloading the images, the notebook will move them to a specified folder in your Google Drive, making it easy for you to access them later. |
|
|
|
|
|
- The default folder is `/content/drive/MyDrive/coco_dataset`, but you can modify the path if needed. |
|
|
|
|
|
## License |
|
|
|
|
|
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details. |
