Sympyのtensor用機能 We have two tensors $latex A \in \mathbb{R}^{I_1 \times \ldots \times I_N}$ and $latex B \in \mathbb{R}^{J_1 \times \ldots \times J_N}$. An element is specified by $latex A(i_1, \ldots, i_N)$, where $latex 1 \leq i_k \l…

scikit-tensor package [code lang="bash"] pip install scikit-tensor [/code]

theano sub-module of theano, used for deep learning. rank: prepared types for rank 0 to 4. higher ranks (>4) tensor must be self-defined. type: dense operations: documentation pytensor the implementation is based on MatLab Tensor Toolbox s…

Blitz++ code: http://sourceforge.net/projects/blitz/ rank: up to 11 type: dense operation: - slicing ITensor well-documented code: https://github.com/ITensor/ITensor rank: type: operation: BTAS code: https://github.com/BTAS/BTAS rank: any …

Tensor Power iteration using Blitz++. The following code is too dirty. [code lang="cpp"] include <iostream> include <blitz/array.h> include <random/uniform.h> include <time.h> include <ctime> include <algorithm> ifdef BZ_HAVE_STD include <fstream> else include <fstream.h> endif blitz::firstIndex i1;…</fstream.h></fstream></algorithm></ctime></time.h></random/uniform.h></blitz/array.h></iostream>

I implemented some functions those are explained in Blitz++ documentation. The code is on my github.

./configure make pdfevince ./doc/blitz.pdf

Blitz++ is a c++ template library for array manipulation. I'm planning to use this library for some tensor computation. Fortunately, we can install the library via Ubuntu package manager as [code lang="bash"] sudo apt-get install libblitz-…