Useful links¶
Document 
Description 

Documentation for programming NumPylike codes on dataparallel devices 

Documentation for programming Numba codes on dataparallel devices as you program Numba on CPU 

Documentation how to manage data and devices, how to interchange data between different tensor implementations, and how to write data parallel extensions 

Performance profiler supporting analysis of bottlenecks from function leve down to low level instructions. Supports Python and Numba 

Analyzes native and Python codes and provides the advice for better composition of heterogeneous algorithms 

Standard for writing portable Numpylike codes targeting different hardware vendors and frameworks operating with tensor data 

Standard for writing C++like codes for heterogeneous computing 

Free ebook on how to program dataparallel devices using Data Parallel C++ 

OpenCl* Standard for heterogeneous programming 

Standard for floatingpoint arithmetic, essential for writing robust numerical codes 

David Goldberg, What every computer scientist should know about floatingpoint arithmetic 
Scientific paper. Important for understanding how to write robust numerical code 
Documentation for Numpy  foundational CPU library for array programming. Used in conjunction with Data Parallel Extension for Numpy*. 

Documentation for Numba  JustInTime compiler for Numpylike codes. Used in conjunction with Data Parallel Extension for Numba*. 
ToDo¶
Todo
Document debugging section