Post date: Sep 15, 2019 3:53:14 AM
Six Important Steps to Build a Machine Learning System
A field guide to thinking about ML projects by Rahul Agarwal.
These are the 6 steps:
Problem Definition
Data
Evaluation
Features
Modeling
Experimentation
The Hitchhiker's Guide to Feature Extraction
Some tricks and code for Kaggle and everyday work by Rahul Agarwal.
Two interesting approaches:
Automatic feature creation using featuretools framework
Using autoencoders
The Five Feature Selection Algorithms every Data Scientist should know
By Rahul Agarwal.
These are the 5 algorithms:
Pearson correlation
Chi-squared
Recursive feature elimination
Lasso from sklearn
RandomForest from sklearn