Kristian Høiem

[AK @_akhaliq]

stable-diffusion-studio: An animation focused workflow frontend for Stable Diffusion

github: https://t.co/7YFbz7xB26

[Biofokus @BioFokus]

FN`s klimapanel sin nye rapport er tydelig på at naturbaserte løsninger er en viktig del av kampen for å redde både klima og natur. Variert natur vil være mer motstansdyktige enn monokulturer mot fremtidige klimaendringer. Vern og beskytt naturskogene!
https://t.co/tqzivBF2lr

[François Chollet @fchollet]

New example on https://t.co/m6mT8SrKDD: timeseries anomaly detection
https://t.co/pK2xRUsmta

[Kevin Markham @justmarkham]

🐼🤹‍♂️ pandas trick:

Want to explore a new dataset without too much work?

1. Pick one:
➡️ pip install pandas-profiling
➡️ conda install -c conda-forge pandas-profiling

2. import pandas_profiling
3. df.profile_report()
4. 🥳

See example 👇

#Python #DataScience #pandastricks

[François Chollet @fchollet]

The most cost-effective way to defeat the Terminators is to introduce a NaN in a SkyNet training data example

[François Chollet @fchollet]

Highway 101, with its soul-crushing traffic jams and millions of tons of CO2 released in the air each year, has less human bandwidth than a single high-capacity subway line in Tokyo

If you think good infrastructure is expensive, wait until you see the cost of bad infrastructure

[Ric Fulop 🇺🇸/acc 🗽 @ricfulop]

1/ New chemistry breakthrough just published by Jeff Dahn. @tesla lead battery researcher. 1M vehicle miles, 795Wh/L, with fast charge and subsequent 20yr durability projected in grid. The data is amazing.
https://t.co/rDLAGcAone

[Siraj Raval @sirajraval]

There's a lot more to machine learning than just selecting the right model! Data cleaning, feature engineering, hyper-parameter tuning, & more are starting to be automated by a few frameworks. I'll explain the growing field of AutoML & compare them here https://t.co/Xfe2ge819e

[Fermat's Library @fermatslibrary]

How do Maxwell's equations predict that the speed of light is constant

[Mark Lynas @mark_lynas]

New Nature paper out today
https://t.co/oAF7nep6ur
states very clearly what is needed if governments are serious about limiting global heating to 1.5C, as agreed at Paris.

Spoiler: No new fossil fuelled infrastructure, anywhere, ever. From now on.

Read thread for details...

[Grant Sanderson @3blue1brown]

The Fourier series video out!

In this animation, each vector rotates at a constant integer frequency. They're added together, tip to tail. The _only_ control you have is the starting position of each, and from that alone, they'll draw almost anything.

https://t.co/1qWWoTPXIA

[Yarin @yaringal]

Really excited to release Bayesian Deep Learning Benchmarks - please share with others who you think might like this, and have a look at the blog/repo/colab:
https://t.co/ZKqN40da1S
This work was done over a period of a year and a half by many collaborators @OATML_Oxford

[Thomas Wiecki @twiecki]

Game changer. Turn your notebooks into interactive web apps with a single command.

[Jeremy Stanley @jeremystan]

1/ I often used data “quality”, “anomaly” and “outlier” interchangeably, but now have a much crisper sense of what I think each term should really mean.

Here are some definitions and advice for what to do about them.

[François Chollet @fchollet]

You don't need to know everything. You don't really need a formal background in this or that -- though it helps. You don't even need a PhD.

You do, however, need to be constantly learning. Be curious. Read books. Don't be "too busy" to learn, or otherwise proud of your ignorance

[Peter Wang 🦋 @pwang]

Panel: Easy interactive dashboards for Python

✅Bokeh
✅Datashader
✅Matplotlib
✅Plotly
✅Altair
✅Jupyter
✅Standalone deployment
✅BSD licensed open source

Great job to the whole team!

[Andrej Karpathy @karpathy]

Speech2Face: Learning the Face Behind a Voice https://t.co/9enUz600fK With increasingly large/effective library of neural net encoders of any X and decoders of any Y, any source of paired data X,Y can give X2Y nets. And opens the door to many X2Y2Z2W...2X

[François Chollet @fchollet]

How to train and evaluate a deep learning model for very imbalanced classification (e.g. classification with 99.82% of negatives and 0.18% of positives)? Here's an example using the Kaggle credit card fraud dataset. https://t.co/HuviYNurim

[Kristian Høiem @KWHoiem]

Hey @elonmusk @SpaceX! Would you consider a nuclear waste disposal system, deploying radioactive waste into deep space?