Welcome to the ulab book!

Introduction

• Introduction
  • Enter ulab
  • Purpose
  • Resources and legal matters
  • Friendly request
  • Differences between micropython-ulab and circuitpython-ulab
• Customising the firmware
  • Compatibility with numpy
  • The impact of dimensionality
  • The ulab version string
  • Finding out what your firmware supports

User's guide:

• ndarray, the base class
  • The ndinfo function
  • Initialising an array
  • Array initialisation functions
  • diag
  • empty
  • Customising array printouts
  • Methods and properties of ndarrays
  • Unary operators
  • Binary operators
  • Comparison operators
  • Iterating over arrays
  • Slicing and indexing
• Numpy functions
  • all
  • any
  • argmax
  • argmin
  • argsort
  • clip
  • convolve
  • diff
  • dot
  • equal
  • flip
  • interp
  • isfinite
  • isinf
  • mean
  • max
  • median
  • min
  • minimum
  • maximum
  • not_equal
  • polyfit
  • polyval
  • roll
  • sort
  • std
  • sum
  • trace
  • trapz
  • where
• Universal functions
  • Computation expenses
  • arctan2
  • around
  • Vectorising generic python functions
• numpy.fft
  • fft
  • ifft
  • Computation and storage costs
• numpy.linalg
  • cholesky
  • det
  • eig
  • inv
  • norm
• scipy.linalg
  • cho_solve
  • solve_triangular
• scipy.optimize
  • bisect
  • fmin
  • newton
• scipy.signal
  • sosfilt
  • spectrogram
• scipy.special
• ulab utilities
  • from_int32_buffer, from_uint32_buffer
  • from_int16_buffer, from_uint16_buffer
• Tricks
  • Use an ndarray, if you can
  • Use a reasonable dtype
  • Beware the axis!
  • Reduce the number of artifacts
• Programming ulab
  • Code organisation
  • The ndarray object
  • Iterating over elements of a tensor
  • Iterating over two ndarrays simultaneously: broadcasting
  • Contracting an ndarray
  • Upcasting
  • Slicing and indexing
  • Extending ulab
  • Creating a new ndarray
  • Accessing data in the ndarray
  • Boilerplate

Indices and tables

• Index

• Module Index

• Search Page