A very Merry Christmas to the entire Data/ML community on Mastodon.Happy holidays!

Any exciting ML topics you plan on deep diving into next year?

A bigger picture is all you need

If y'all are free tonight @ 7pm ET, y'all should check this out.

Even if you're not in AI, the way it is growing chances are you'll be somewhat involved in AI at some point.

It's always a good idea to get informed on these kind of topics.

#100DaysOfMLCode
#DEVCommunity
---
RT @rajiinio
Tonight, I'll be giving a talk on "The Challenges of Audits, Accountability & Algorithmic Justice".

I won't be talking much about solutions. Instead…
https://twitter.com/rajiinio/status/1346824910296010757

Very interesting article from @jetbrains

They analyzed 10,000,000 notebooks and share the trends.

This is mainly aimed at Data Science and ML.

If you're wondering which tools are mostly used, check the article out

#100DaysOfMLCode

https://blog.jetbrains.com/datalore/2020/12/17/we-downloaded-10-000-000-jupyter-notebooks-from-github-this-is-what-we-learned/

Software Developer Secret

Is there's a specific issue you've been having a hard time debugging? Add better logging.

Use logging to show the state of data before the error shows up.

Check this out!

Statistics concepts explain with visualizations.

I'm on chapter 3 "probability distributions" (has been my nemesis since college) and it's amazingly digestible!

By @SeeingTheory

And they're writing a book, what?!

#100DaysOfMLCode

https://seeing-theory.brown.edu/

If you were thinking how can LDA be improved, checkout this thread and the links in it.

It's a really good idea.

#100DaysOfMLCode
---
RT @dam_nlp
The third paper in 14C:

Topic Modeling in Embedding Spaces

Adji Bousso Dieng (@adjiboussodieng), Francisco Ruiz, David Blei

TACL: https://www.mitpressjournals.org/doi/full/10.1162/tacl_a_00325
Talk: https://slideslive.com/38939405

#EMNLP2020
1/
https://twitter.com/dam_nlp/status/1329121755655499776

RT @iamtrask@twitter.com

Machine Learning in a company is 10% Data Science & 90% other challenges

It's VERY hard. Everything in this guide is ON POINT, and it's stuff you won't learn in an ML book

"Best Practices of ML Engineering"

This is a lifesaver #100DaysOfMLCode project

http://martin.zinkevich.org/rules_of_ml/rules_of_ml.pdf

#100daysofMLcode
#day05 I didn't do much. I got busy with setting up new gitlab account and getting familiarised with it. Did some assignment work.

#competitiveprogramming solved a problem set and uploaded in my gitlab account.

https://gitlab.com/ScienceSaint/problemset-and-solution

Feel free to comment and coming up with optimal solution for the problem set.

#100daysofMLcode
#day04 completed multiple linear regression and backward elimination method.

#competitiveprogramming solved a problem based on math and binary search.

#100daysofMLcode
#day03 learnt about multiple linear regression and how to find out the significant independent variables and remove insignificant variables through backward elimination to increase the efficiency of the model.

#competitiveprogramming didn't solve a problem. Slept due to tiredness.

#100daysofMLcode
#day02 learnt about multi variable linear regression and math behind the linear regression algorithms.

Started to focus on competitive programming besides machine learning to increase problem solving skills.

I am using codeforces.com to increase my problem solving ability in short period of time and using codechef.com to increase my problem solving ability in long period of time.

Aim is to solve one problem each day.
Solved a problem based on binary tree and divide&conquer.

Started my journey to learn machine learning.
#100daysofMLcode
#day01 learnt data processing in python and simple linear regression algorithms.