AIscm Reference Manual

This manual is for AIscm (version 0.25.2).

AIscm is a Guile extension for numerical arrays and tensors. Performance is achieved by using the LLVM JIT compiler.

Documentation

• Installation
• Built-in assembler
• Input/output
• Arrays
• Array operations
• Convolutions
• Tensorflow bindings

See also

• Github page
• Gitlab page
• Bitbucket page
• Initial package review email thread
• Debian package tracker
• Debian build logs
• Ubuntu Launchpad entry

Credits

• Jan Wedekind: software development
• Andrey Rahmatullin: Debian package sponsor
• Gianfranco Costamagna: Ubuntu package

External links

• LLVM ORC JIT C API
• Chainer: A Powerful, Flexible, and Intuitive Framework of Neural Networks
• DyNet: dynamic neural networks library
• Torch: Scientific computing for LuaJIT
• Theano: Python library for JIT compiled array operations
• PyTorch: Tensors and dynamic neural networks in Python with strong GPU acceleration
• Tensorflow: numerical computing with dataflow graphs
• tensorflow.rb
• Numpy: Python library implementing n-dimensional arrays
• Bergstra et al.: Theano: A CPU and GPU math compiler in Python
• Lush: large-scale numerical computing
• Vigracket: Vigra bindings for Racket
• Ceemple C++ OpenCV IDE
• Arrayfire: accelerated computing library
• Halide: a language for image processing and computational photography
• PeachPy: Portable assembly for Python
• Guile manual
• Guile-CV: computer vision library functional programming library for Guile
• GOOPS: object-oriented extension to Guile
• COOPS: Chicken Scheme object system
• System V Application Binary Interface
• Intel® 64 and IA-32 Architectures Software Developer Manuals
• CLASP: Lisp with LLVM Jit and C++ foreign interface
• Embedded Language for Accelerated Array Computations
• SymPy: computer algebra in Python
• Maxima computer algebra system
• Julia language
• Kona language
• Guile SDL2
• Guile OpenGL
• Jakobsen: Advanced Character Physics
• COMP768: Rigid Body Dynamics (I) (I)

Copyright © 2013, 2014, 2015, 2016, 2017, 2018, 2019 Jan Wedekind.