⌛️ Save tons of time debugging machine learning models.
✅ Use overfitting to detect bugs quickly.
👩🏽💻. A concrete code example with PyTorch and Lightning
Debugging machine learning models can be hard and time consuming. Many thing can go wrong:
A bug in data loading
A bug in the mode itself
A bug in the training loop
...
As bad as overfitting sounds, it can be your ally to debug machine learning models quickly. Let's see how!
👉🏽 Link to the demo Notebook: https://github.com/bitswired/bitsof-a...
Hope you'll enjoy it!
🌐 Visit my blog at: https://www.bitswired.com
📩 Subscribe to the newsletter: https://newsletter.bitswired.com/
🔗 Socials:
LinkedIn: / jimi-vaubien
Twitter: / bitswired
Instagram: / bitswired_
00:00 Intro
00:21 What Is Overfitting?
01:53 How Overfitting Helps Debugging?
02:58 Code Demo Setup
03:24 Debug With Overfitting In PyTorch
06:46 Outro
Resources:
PyTorch Lightning Trainer can help you overfit a small amount of data to debug your model: https://pytorch-lightning.readthedocs...
Overfitting: https://en.wikipedia.org/wiki/Overfit...
On this page of the site you can watch the video online 🔍 Debug ML With Overfitting: PyTorch Lightning (Tutorial + Example) with a duration of hours minute second in good quality, which was uploaded by the user Jimi V. (Bitswired) 31 May 2022, share the link with friends and acquaintances, this video has already been watched 1,124 times on youtube and it was liked by 24 viewers. Enjoy your viewing!