(category for greek, keywords, sentences, lemma/tags on text results)

Make sure spaCy models (and GR classification model) are loaded only when necessary

• upload Greek fastText model that was produced
• add instructions on how to produce/use for other languages

• tests

• flake

• pep8

eg when you hover over a NER, on the right side, highlight on text

Add instructions on how to install + test environment Tested on Ubuntu 16.04 with python 3.5 create virtual env: virtualenv -p python3.5 . source activate: source bin/activate clone project: https://github.com/eellak/text-analysis/ enter project dir: cd text-analysis install requirements: pip install -r requirements.txt run django migrations: python manage migrate install nginx: apt-get install nginx setup nginx + uwsgi configs: deploy/ ... . make sure to set the servername visit url...

On scrolling the website, a back-to-top button can be included in the bottom-right corner to make it easier to navigate back to the top.
Please let me know if you need any help with this or if I can integrate this in the existing UI.

a more light way as displaCy does it would also be beneficial on the sentences section

improve summarization (currently stub)

the way it does it upon start

just tried with https://github.com/saffsd/langid.py

also check fastText: https://fasttext.cc/blog/2017/10/02/blog-post.html

this could be a switch that shows lemmatized sentences, inside the sentences section (also inside the same structure on the API response)