NumPy

Fast math on whole grids of numbers. The engine under pandas and machine learning. 15 modules of 10 minutes each, with questions that check your answers in the browser.

Start the track

Foundations

• 1. Why NumPy: arrays vs Python listsFree
  Meet the array, the data structure behind pandas and machine learning, and see why it beats lists for numbers.
• 2. Creating arrays (array, arange, zeros, ones, linspace)Free
  Build arrays from lists, number ranges, evenly spaced points and filled grids, the starting point of every NumPy program.
• 3. Array attributes (shape, dtype, ndim)Free
  Read an array's shape, size and element type, convert between types, and avoid silent overflow.
• 4. Indexing and slicingFree
  Pick single values, ranges, rows, columns and sub-grids out of arrays, and change them in place.
• 5. Boolean and fancy indexingFree
  Filter arrays with conditions, combine conditions safely, replace values that match, and pick elements by a list of positions.

Intermediate

• 6. Element-wise operations and broadcastingFree
  Combine arrays of different shapes without loops, using NumPy's broadcasting rules and universal functions.
• 7. Aggregations (sum, mean, and the axis argument)Free
  Summarise arrays as a whole, per row or per column, find where the maximum is, and handle missing values.
• 8. Reshaping, stacking, and splittingFree
  Change an array's shape without changing its data, glue arrays together, and cut them into pieces.
• 9. Random numbers and seedsPro
  Generate reproducible random data, sample and shuffle, and use simulation to answer "what are the chances?" questions.

Advanced

• 10. Linear algebra basics (dot, matmul)Pro
  Compute weighted sums, order totals and model predictions with the dot product and matrix multiplication.
• 11. Working with axes in multi-dimensional arraysPro
  Handle 3-D data like stores × months × products, aggregate over several axes, and reorder dimensions.
• 12. Views vs copies (and why it matters)Pro
  Know when an array shares data with another, avoid functions that secretly change their input, and use in-place operations on purpose.
• 13. Vectorization: replacing loopsPro
  A toolkit of patterns for turning Python loops into fast array operations, from if/else chains to group sums and nearest neighbours.

Expert

• 14. Structured arrays and dtypesPro
  Store records with named fields of different types, load them from files, and work with dates and business days.
• 15. Performance and memory layout (contiguity)Pro
  Understand how arrays sit in memory, why row order and strides affect speed, and how to avoid unnecessary copies and temporaries.

Track project

• NumPy Project: Toko Kita's pricing and planningPro
  Three graded tasks with arrays and no loops. Price a promotion on every order line at once, compare three stores month by month with axes and broadcasting, and simulate a year of demand to see how often a product would run out.

Pro modules unlock with one payment of Rp 149.000, yours to keep.