Datanatomy
VideosBlogCodeAboutLinks
/

    GitHub
    DatanatomyCode is MIT on GitHub
    YouTubeTikTokInstagramThreadsGitHubRSS
    Datanatomy
    VideosBlogCodeAboutLinks
    /

      GitHub
      14 videos

      Machine learning, taken apart.

      Long explainers, 30-second cuts and quick tips. Open one for the player, the key moments, the full transcript and the code.

      All 35Machine learning 14AI engineering 4Data engineering 11Methodologies 4Software engineering 2

      1–12 of 14 vertical videos

      Series
      Length
      Watch on
      Sort
      • Frame from How to choose a model: Choosing a model, data first, goal second, model last2:26Machine learningHow to choose a model
      • Frame from Overfitting explained: Overfitting, why a perfect score is a red flag2:25Machine learningOverfitting explained
      • Frame from A neural network by hand: Neural network, from scratch, NumPy only2:28Machine learningA neural network by hand
      • Frame from Decision trees explained: Decision trees, twenty questions, from scratch2:20Machine learningDecision trees explained
      • Frame from Always stratify a rare class: Stratified split, keep the rare class in your test set0:21Quick tipAlways stratify a rare class
      • Frame from Linear regression, no libraries: Linear regression, from scratch, no libraries1:23Machine learningLinear regression, no libraries
      • Frame from 80% accuracy, zero churners caught: Choosing a model, goal first, model second0:36Machine learning80% accuracy, zero churners caught
      • Frame from A perfect score is a red flag: Overfitting, why a perfect score is a red flag0:39Machine learningA perfect score is a red flag
      • Frame from Why one neuron can't solve XOR: Neural network, XOR in 9 lines, NumPy only0:39Machine learningWhy one neuron can't solve XOR
      • Frame from The model you can actually read: Decision trees, twenty questions, 9 lines0:38Machine learningThe model you can actually read
      • Frame from 9 lines of Python that learn: Linear regression, 9 lines, no libraries0:35Machine learning9 lines of Python that learn
      • Frame from K-means: groups with zero labels: K-means clusters, finding groups nobody labeled0:34Machine learningK-means: groups with zero labels
      ← Previous
      1. 1
      2. 2
      Next →
      DatanatomyCode is MIT on GitHub
      YouTubeTikTokInstagramThreadsGitHubRSS