|
Description:
|
|
Machine learning and data science are full of best practices and important workflows. Can we extrapolate these to our broader lives? Eugene Yan and I give it a shot on this slightly more philosophical episode of Talk Python To Me.
The seven lessons:
1. Data cleaning: Assess what you consume
2. Low vs. high signal data: Seek to disconfirm and update
3. Explore-Exploit: Balance for greater long-term reward
4. Transfer Learning: Books and papers are cheat codes
5. Iterations: Find reps you can tolerate, and iterate fast
6. Overfitting: Focus on intuition and keep learning
7. Ensembling: Diversity is strength
Links from the show
Sponsors
Retool Linode Talk Python Training |