Practical guides to learning AI and machine learning, written by Gaurav Jain. Worked examples, projects with complete code and learning plans you can follow.
Data leakage is when a model learns from information it won't have at prediction time. See target, preprocessing and time leakage examples, and fixes.
troubleshooting9 min read
Deep learning is machine learning with many-layered neural networks. Compare data, features, hardware and real examples, and see when to use each.
comparison12 min read
Tell overfitting from underfitting with training and validation error, work through a demand example and choose a sensible next experiment.
troubleshooting6 min read
Our advice: learn PyTorch first, unless your course, team or project uses TensorFlow. See the same model in both, a side-by-side table and a decision rule.
comparison10 min read
Fine-tuning is one kind of transfer learning: it updates pretrained weights, while feature extraction freezes them. When to use each, with PyTorch code.
comparison9 min read
Misleading statistics use true numbers to suggest false conclusions. See 10 examples, from truncated axes to Simpson's paradox, and how to spot each one.
explanation16 min read
The statistics and probability data science and machine learning use, in 7 steps, with a worked example for each and a list of what you can skip.
learning path9 min read
Runnable machine learning sample code in Python: a scikit-learn classifier explained line by line, plus regression and clustering, each with its output.
tutorial11 min read
The prerequisites for machine learning: Python, basic linear algebra and calculus, and intro statistics and probability. Exact topics and a readiness test.
learning path11 min read
The courses take each idea from the intuition and the maths to code you write yourself, from your first line of Python to machine learning.