This system fits the '67% size' in 'Chrome Browser'. In other sizes or browsers, some contents may not be displayed as intended.
1.Activate your virtual environment
2.Install all the required packages through 'pip install -r requirements.txt' in the directory
3.Build mongodb environment (If you want to use the system as cloud, you should use mongdb atlas)
- If you want to change adjectives, revise the 'CONST_ADJECTIVE'
- If you want to change numbers, revise the 'CONST_BLUE_NUMBER, CONST_RED_NUMBER, CONST_NEUTRAL_NUMBER'
- If you want to use MongoDB Atlas, direct your url in the client = pymongo.MongoClient("your_url")
- If you want to use your own images, you can add them ./app/static/image/final_dir. Put this directory in CONST_IMAGE_PATH starting from static/
- If you want to change any parameters including but not limited to 'keyword, images, user_id, numbers, labels, attributes, clusters, and etc', you need to revise codes in routes.py, photolabeling.html, main.js where parameters are handled
- You must extract the feature vectors of images at first trial using the codes at route.py line 45 ~ 66, Please use image feature vectors from a saved pickle file.
"# photo-labeler"