[WARNING]

• This repository is now archived.
• Please check the Kotlin implementation, FlickrDataset.

flickr-dataset

Simple python script for downloading massive images from flickr.

Functions

• Download images by keyword from flickr (command: download).
• View individual images, select images to delete (command: check).

How to use

Setup python environment

1. Clone this repository and setup virtual python environment.
2. Install packages with pip.

   pip install -r requirements.txt
   

3. Copy config.example.json as config.json and fill out.

Configuration File

{
  "API_KEY": "[YOUR FLICKR API KEY]",
  "API_SECRET": "[YOUR FLICKR SECRET KEY]",
  "KEYWORD": "kitten",
  "SIZE": 128,
  "START_PAGE": 1,
  "END_PAGE": 10
}

The SIZE should be bigger than or equal to 128, smaller than or equal to 640.

Download images massively

python main.py download

approx. (END_PAGE - START_PAGE + 1) * 100 images will be downloaded.

Check images one by one

python main.py check

The Qt window will show up, and images located in tmp_{KEYWORD} will be shown on the window one by one.

There are four buttons available.

• Mark/Unmark: Mark or Unmark image.
• Prev: Show previous image.
• Next: Show next image.
• Stop: Stop, and remove all marked images.