Since OpenPaths is dead, we switched to WHIB, which isn't owned by Facebook (or actually also dead now) and doesn't drain the battery of my phone quickly.
A 2$ premium feature let's you export all the data as CSV, which is easy to parse and display.

whib

1aff53abca53a627bb4452e60986bcc9_preview.png

6639a93c6c5ab661a0623db4d67794a3_file-kLwO5UDauO-profile.jpg

In this notebook I would like to share how to process the work with text to build topics. To make it more attractive - apply NLP to the Stock Descriptions.
We will load and build the models with cool 3d visualization.
I encourage you to clone this notebook and play with every parameter to fine tune the result as your sticky mind would like.

nlp-plot-in-3d-stocks-description

24922b95e2c0e8c1b813688b8f2fa83f_Screenshot3.png

parsing-sample

26258e5d189e859e00241c01020d4106_preview.png

parse-studies

dd7f4ba5b024104838829e04c6cc2960_preview.png

kyso-tips-and-tricks

17002043ebb58d3b154caccaf9d1ac8e_preview.png

139f9e5815fae4b206cc783868079511_file-CP5n4TTHy7-profile.jpg

ml-with-python

056bb79e0ba9208b9db8469edd15d6da_preview.png

regularized

1feedabcc772ae3f02482acc853f1f18_preview.png

logreg

0d145bc06d3ed6a40479faf343aafe63_preview.png

Kyso