Deep Learning Project on Comics
- main.py : wrapper for pythonscraper
Usage : python3 main.py [db|metadata|download]
db : creates and fills the databas according to comics.json queries
metadata : downloads metadata from comicvine for the comics in the db
download : downloads a pseudorandom amount of files (issues with comicvine metadata)
-
pythonscraper : contains files for scraping comicextra.com, comicvine.gamespot.com and store comic images and their metadatas in a sqlite database
-
supervised_learning : Uses fine tuning to learn Date deatures from the downloded comics
-
unsupervised_learning : Uses An AutoEncoder + KMeans Clustering for classifying the comic images
-
comics : folders and comic images
-
comics.db : prescraped database (done with pythonscraper)
-
monitor.sh : monitor the downloads in the comics folder