Pixel Process
  • Projects
  • Workflow
  • Foundations
  • About
  • Contact
  1. Build Models
  • Foundations
  • Try Python
    • Datatypes and Operators
    • Variables and Functions
    • Iteration and Flow Control
    • Errors and Experimentation
  • Explore Data
    • Hello World
    • Expert Debugger
    • Dataset Basics
    • Visualization Basics
    • Visualization Building Blocks
    • Image Basics
  • Build Models
    • Combining Colors: Sampling Analysis
    • Random Forest: Deep Dive

Build Models

ML Workflows in Full Python Environments

End-to-end machine learning notebooks running in complete cloud environments via Binder. Full package access — scikit-learn, numpy, pandas, matplotlib — in the same kind of environment you’d use for real work.

WarningBefore You Start
  • Launch time: Binder environments take 30 seconds to 2 minutes to spin up. This is normal.
  • Session timeout: Environments shut down after ~10 minutes of inactivity. If it times out, relaunch from here.
  • Saving work: Your work is not saved on the server. Download any notebooks you want to keep before closing the tab (File → Download as → Notebook).

The friction is intentional — these are real compute environments, and working with them is part of the skill.


Notebooks

Combining Colors: Sampling Methods How different sampling strategies affect outcomes. An interactive demonstration using colored samples that builds intuition for why sampling decisions matter before you ever touch a model.

Launch in Binder →

Random Forest Ensemble methods from the ground up. How random forests work, when they’re the right choice, and how to diagnose their behavior through feature importance and error analysis.

Launch in Binder →


What You’ll Need

Comfort with pandas and basic data manipulation. If you haven’t worked with dataframes yet, start with Explore Data first.

Built with Quarto

Pixel Process (PxP) | Think Clearly | Build Carefully | Apply Rigorously

  • Dexterous Data Consulting