If you want to go really deep on this, check out Stan, which is a whole language for probabilistic programming. Here's some survival models for Stan:https://t.co/DPFLm2bbzO
— Jeremy Howard (@jeremyphoward) March 12, 2020
If you want to go really deep on this, check out Stan, which is a whole language for probabilistic programming. Here's some survival models for Stan:https://t.co/DPFLm2bbzO
— Jeremy Howard (@jeremyphoward) March 12, 2020
What's even more awesome is that @AllenDowney has made the chapters available as runnable colab notebooks, which is just about the best way to learn I know of!https://t.co/vjM9x0AU24
— Jeremy Howard (@jeremyphoward) March 12, 2020
Most decision tree ensemble libs can do survival analysis. Here's a nice intro for the wonderful xgboost lib: https://t.co/LfFEPjmcpt
— Jeremy Howard (@jeremyphoward) March 12, 2020
There are *lots* of ways to model censored data. Here's a great little summary of a few, which shows (for 2 datasets) that random forests (with some tweaks) works best.https://t.co/VhQYiaT79Q
— Jeremy Howard (@jeremyphoward) March 12, 2020
Survival analysis is used to handle "censored data". That means: data where we sometimes don't know the label yet.
— Jeremy Howard (@jeremyphoward) March 12, 2020
Most data scientists aren't familiar with this technique. That's understandable, because there's lots of fields where it doesn't really come up. But we need it now! pic.twitter.com/mHjIz1lLw7
Kaggle Grandmaster @vigloviko provides an in-depth look into his team's experience hosting the Lyft 3D Object Detection for Autonomous Vehicles competition, along with a break down of the winning solutions 🏅|| [READ] https://t.co/OpydqZERMa
— Kaggle (@kaggle) March 12, 2020
Curios how does the state of the art image classification training pipeline looks?
— Vladimir Iglovikov (@viglovikov) March 11, 2020
Check this video by @kaggle Grandmaster Arthur Kuzin.
"Bag of Tricks for Image Classification".
Slides: https://t.co/7Wg7uCYne3
Presentation: https://t.co/pAeVGF7ENP
[WATCH] @rctatman break down how to store your keys, secrets, and tokens when using a Kaggle notebook! https://t.co/JHhzOGYdpH
— Kaggle (@kaggle) March 10, 2020
NEW: I've made a Colab specifically for the "stable colorizer" as another option for you guys when using the open source version of DeOldify. Main differences- it does skin better and tends to have less glitches, but produces less interesting colors. https://t.co/f1kypZQ4re
— Jason Antic (@citnaj) March 10, 2020
PyTorch + Cloud TPU + Colab: a set of code pointers and notebooks to get you started. https://t.co/HhG8vX3Q2R
— PyTorch (@PyTorch) March 10, 2020
Zoom In: An Introduction to Circuits - A new Distill article by @ch402, @nicklovescode, @ludwigschubert, @gabeeegoooh, @mpetrov and @shancarter https://t.co/8gScfsj8rQ
— Distill (@distillpub) March 10, 2020
From PyTorch to JAX: towards neural net frameworks that purify stateful code: https://t.co/Sg3k4XpzTD
— Denny Britz (@dennybritz) March 10, 2020
Great writeup from @sjmielke on how to think about JAX programs and how it all works - from scratch.