PV254-recommender-system

We had some problems with a image uploading to Heroku so during evaluation of a project it is hosted here.

Cloud Vision API dependencies

env variable GOOGLE_APPLICATION_CREDENTIALS needs to be set to the path where service account key is stored
it is sufficient to place access.json inside .secrets folder

Install virtualenv via pip install virtualenv (alternatively pip3 install virtualenv).
Create new virtual environment virtualenv venv (alternatively virtualenv -p python3 venv).
Activate virtual environment source venv/bin/activate (Windows: venv\Scripts\activate).
Install all requirements via pip install -r requirements.txt
Leaving virtual environment: deactivate.

Export PYTHONPATH to root dir export PYTHONPATH=`pwd` .
Download random images python scripts/download_random_photos.py.
Download Instagram profile python scripts/download_ig_profile.py --profile jpancik.
Annotate random images python scripts/annotate_images.py --input resources/random-images/.
Annotate IG profile images python scripts/annotate_images.py --input profiles/jpancik/.
Rank images python scripts/rank_images.py --library profiles/jpancik/ --candidates random-images/.
Open HTML with results in resources/random-images/recommendations.html.
$$$ Profit $$$.

Perform all steps in Setup
Run server:
- dev python manage.py runserver is listening on 127.0.0.1:8000
- prod gunicorn --timeout 300 frontend.wsgi

python scripts/export_label_count_from_profiles_to_csv.py --dbfile <PATH_TO_DB_FILE> --output <PATH_TO_EXPORT_CSV> --min_score <OPTIONAL_MIN_SCORE>

Download model from https://www.dropbox.com/s/buix0deqlks4312/lexvec.commoncrawl.ngramsubwords.300d.W.pos.bin.gz?dl=1
Unzip it to /lib/lexvec_model/lexvec_model.bin