1st session in English: all activities accounted for in rating

Aditya Soni edited this page Oct 15, 2018 · 16 revisions

This are all activities accounted for in the 1st session of the course (Feb. 5 – Apr. 26, 2018).


Announced in the #eng_mlcourse_open channel in OpenDataScience Slack team (fill in this form to join). Check Pinned Items in this channel. Also announced in articles on Medium.



You’ll try to identify a user on the Internet tracking his/her sequence of attended Web pages. The algorithm to be built will take a webpage session (a sequence of webpages attended consequently by the same person) and predict whether it belongs to some Alice or somebody else.


  • the maximum number of submissions per day is 5, this is an individual competition, i.e. max team size = 1, team merges are not allowed
  • if you’d like to get credits in our course rating, then rename your team on Kaggle in full accordance with your name in the rating (the rating will appear pretty soon after the 1st assignment’s deadline)
  • up to April 15 (inclusive), different benchmarks may appear, and credits are awarded only to those who beat all benchmarks on private LeaderBoard (ROC AUC is strictly greater than that of all benchmarks)
  • if you overfitted and suffered from shake-up (so you’ve beaten all benchmarks on public LB but failed to beat all benchmarks on private LB ) – then no offense, there’ll be no credits
  • deadline is April 23, 2018
  • after the deadline, till April 25, 2018 (23.59 CEST), those who beat all benchmarks should upload their reproducible solutions (python scripts) here
  • the competition finals as well as the final rating of the course will be published on April 27


In this competition your task will be to predict the number of "recommendations" for an article published on Medium.

Rules are the same as for Alice competition, the only difference is the link to upload solutions. And in #mlcourse_open, you can get credits for only one competition that you choose.

Credits for competitions in our course:

  1. To get credits you have to:
    • Beat "Assignment 6 baseline" on private leaderboard (solutions for A6 are not released until the end of the course)
    • Upload your reproducible solution till April 25 (23.59 CEST)
    • Rename your Kaggle team (consisting of one person) in full accordance with your name in the course rating
    • 1 place - 40 credits
    • 2 place - 30 credits
    • 3 place - 25 credits
    • 4-10 places - 20 credits
  2. The final formula for credits in both competitions is the maximum among credits for Alice and for Medium

Kaggle kernels

Additional credits are distributed for nice EDA (Exploratory data analysis) in our competitions. Form it as a Kaggle kernel, publish here or here (Alice or Medium) It is possible to include training some models, but only so that the results are not too good: Alice: AUC < 0.95. Medium: MAE > 1.5. Vote for nice kernels on Kaggle itself (upvote in the upper right corner). The results will be counted in the end, on the 22nd of April.

  • 1-st place - 10 credits
  • 2-nd place - 7 credits
  • 3-rd place - 4 credits
  • 4-th place - 2 credits
  • 5-th place - 1 credits

For each competition.

Requirements (once again):

  • team name should be coincide with the name in our course rating
  • results on LB should not exceed the above thresholds, without spoilers plz, these competitions will still be used in future
You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.
Press h to open a hovercard with more details.