Hi,
DSB is incoming, weekend is incoming and in Prague it’s probably raining so let’s start with reading! Very interesting article is the first one from Analytical about definition of classic literary work.
As always, enjoy your reading.
Analytical
https://post45.org/2021/04/the-goodreads-classics-a-computational-study-of-readers-amazon-and-crowdsourced-amateur-criticism/ – This is an amazing analysis that is trying to define a classic literary work.
https://nvlabs.github.io/GANcraft/ – GANcraft is an unsupervised neural rendering framework for generating photorealistic images of large 3D block worlds such as those created in Minecraft. (rcmd by reader)
https://www.quantamagazine.org/new-neural-networks-solve-hardest-equations-faster-than-ever-20210419/ – Deep neural networks solve partial differential equations faster than you might think.
Computer Science & Science
https://github.com/git-tips/tips – Wow! List of git tips! Very handy! (rcmd by reader)
https://mcfunley.com/choose-boring-technology – Boring technologies have advantage, they’re well understood. Don’t overdo your technology stack. (rcmd by reader)
https://blog.gitguardian.com/safely-open-source-software-best-practices/ – What is necessary to do when you are providing your internal project as an open source.
Graphs and Visualizations
https://www.lorismat.com/work/football – Well, this is a nice visualization of american football salaries but quite selfserving to be honest. (rcmd by reader)
https://www.analyticsvidhya.com/blog/2021/04/pandas-visual-analysis-interactive-visual-analysis/ – When you need a quick look on data use Pandas Visual Analysis.
https://www.analyticsvidhya.com/blog/2021/04/animated-bar-graph-data-science-project/ – Bar chart races are popular. You can create one with online tools like Flourish or use bar_chart_race library in Python.
Business and Career
https://svpg.com/revenge-of-the-pmo/ – Agile is just a disguise waterfall, or at least so if you don’t move from project-mindset to product-mindset. (rcmd by reader)
https://benjiweber.co.uk/blog/2021/04/10/dont-hire-top-talent-hire-for-weaknesses/ – Who to hire in order to make a team stronger?
https://www.kdnuggets.com/2021/04/consider-being-data-engineer-instead-data-scientist.html – Everybody should be or aim to be a data engineer.
Pop
https://digital-strategy.ec.europa.eu/en/library/proposal-regulation-laying-down-harmonised-rules-artificial-intelligence-artificial-intelligence – Proposal for a regulation of AI in the EU. Long document but it might change the whole game. (rcmd by reader)
https://9to5google.com/2021/04/27/mighty-browser/ – Mighty streams browser from a cloud. It seems like the only way to make Chrome fast.
https://www.vice.com/en/article/k78a53/the-irs-wants-help-hacking-cryptocurrency-hardware-wallets – The IRS wants to break into cryptocurrency hardware wallets.
Education
https://www.reddit.com/r/learnprogramming/comments/m8nmuu/all_the_mooc_of_helsinki_university/ – Free MOOC from Helsinki University and Aalto University and some of these courses look more than good! (rcmd by reader)
https://evidentlyai.com/blog/tutorial-2-model-evaluation-hr-attrition – Tutorial on evaluation ML models, how to compare them and choose the best one.
https://www.kdnuggets.com/2021/04/production-ready-machine-learning-nlp-api-fastapi-spacy.html – How to implement an API based on FastAPI and spaCy for Named Entity Recognition. Or for change another “api” article about connecting DataBricks and MongoDB with Python API.
Data & Libraries
https://www.analyticsvidhya.com/blog/2021/04/automate-nlp-tasks-using-evalml-library/ – EvaIML is an AutoML library, in this article used for NLP.
https://gradientflow.com/what-is-dataops/ – Thoughts on DataOps.
https://github.com/ShopRunner/collie_recs – A library for preparing, training, and evaluating recommender systems using PyTorch. And at least on paper it seems good.
Video & Podcast
https://riveducha.onfabrica.com/openai-powered-linux-shell – Have a look on the Linux shell powered by AI!
https://www.youtube.com/watch?v=r7SgSegtfK0 – MLOps in AWS, a nice video about basic principles. And here you can read what is wrong with MLOps.
Papers & Books
https://riveducha.onfabrica.com/openai-powered-linux-shell – Have a look on the Linux shell powered by AI!
https://www.notion.so/Paper-Notes-by-Vitaly-Kurin-97827e14e5cd4183815cfe3a5ecf2f4c – Vitaly Kurin is a PhD student at the University of Oxford working on Multitask Reinforcement Learning in Graph-Based and he takes notes on papers he read and put them online.
Behind the Fence
https://www.tecton.ai/careers-all/?gh_jid=4757867002 – Software engineer in Tecton, San Francisco or New York, USA.
Joke
https://nostalgebraist.tumblr.com/post/649233680736337920/the-scikit-learn-cargo-cults – Epic! 😀 “One cannot rule it out – that data scientists do not know how to do anything other than type the words “fit” and “predict.”” (rcmd by reader)
https://img.devrant.com/devrant/rant/r_75099_JkRnz.gif – What it’s like to debug code? 😀
[…] and unpleasant reading, but Ada Lovelace Institute tackles and critizes the AI Act by EU mentioned in DSB #115 that will affect the whole data science in EU and maybe even […]