David Gasquez @davidgasquez@mastodon.xyz
Pinned toot
If you wish to make an apple pie from scratch, you must first invent the universe — Carl Sagan
The Government Protects Our Food and Cars. Why Not Our Data? https://buff.ly/2NFK1al
RT @iridium77@twitter.com
Python 3 and Python streaming on Dataflow are generally available! Thank you all for making it happen!
Effective Study Skills for College Students:
Asking “Why?” Questions
https://buff.ly/2WDYIh3
This was really unexpected! Which Planet is Closest?
"There's no love in a carbon atom, No hurricane in a water molecule, No financial collapse in a dollar bill." – Peter Dodds
The problem with metrics is a big problem for AI
http://fast.ai https://buff.ly/31t16bi
An Interactive Introduction to Fourier Transforms https://buff.ly/2MiuZql
RT @jmgilgado@twitter.com
I’ve just published a guide to be a better remote worker:
https://josemdev.com/articles/how-to-be-a-better-remote-worker/
What My Data Science Degree Did Not Teach Me — Part 1.
What are some websites like https://distill.pub? Looking for distilled and beautifully presented papers on any topic. 📚
RT @cathalhoran@twitter.com
The range of #NLP features and how easy they are to use in @spacy_io@twitter.com is really impressive. Great overview of a few here along with some useful spaCy add-ons such as language detection https://medium.com/eliiza-ai/a-tour-of-awesome-features-of-spacy-part-1-2-58b32425954f via @EliizaAI@twitter.com
A/B testing helps organizations make decisions that lead to increases in conversion, engagement, and other metrics, but it is not a silver bullet.
The State of Machine Learning Frameworks. PyTorch has the research market, TensorFlow the industry.
You never 'win' against entropy; you just try to last as long as you can.
Congratulations for 200 awesome issues of the DSR @jthandy@twitter.com! Thanks for doing the best Data Science/Engineering newsletter. 🎉
RT @jeremyphoward@twitter.com
After completing first 3 lessons of https://course.fast.ai, @vedrangrcic@twitter.com completed the assignment using a dataset of clouds. He shared it on the forums.
Then it got re-trained on world's largest cloud dataset and is now deployed in 120 countries!
Fantastic work, Vedran! :)
RT @stephen_wolfram@twitter.com
The $30,000 question: how random is rule 30? Will we know this year, this decade, this century? (For the digits of pi, it's been 100+ years…) https://wolfr.am/H05meab8
What’s a good learning rate? How many hidden layers should your network have? Is dropout actually useful? Why are your gradients vanishing?
Check this awesome guide by @lavanyaai@twitter.com to help you gain some intuition around Neural Networks.
https://towardsdatascience.com/designing-your-neural-networks-a5e4617027ed
RT @tobi@twitter.com
Tim asked me on his podcast "If you could put a billboard up that everyone would see, what would it say?" I didn't have a good answer back then, but if he'd ask me now I'd say "Go read every word of @waitbutwhy@twitter.com's series The Story of Us" https://waitbutwhy.com/2019/08/story-of-us.html
RT @dask_dev@twitter.com
Many Scikit-Learn workflows can be easily parallelized with Dask.
Operations like grid search, random forest, and others that use the njobs parameter in Scikit-Learn can automatically hand-off parallelism to a Dask cluster.
