Marc Glisse
Marc Glisse
Hello, my goal is to work around https://github.com/pytorch/pytorch/issues/34452 . ```python import torch import numpy as np import eagerpy as ep t=ep.astensor(torch.tensor([0.])) i=np.array([[0,0],[0,0]]) t[i] t[ep.astensor(i)] ``` Currently, the last 2 lines...
Hello, with a pytorch tensor t, I can call `t.norm(p, dim)`. This gives a similar result to eagerpy's `lp`, but makes a huge difference when it comes to the gradient....
## Issue Details This test fails because CGAL/float.h and CGAL/int.h are not included. When GMP is present, it includes number_type_basic.h, which in turn includes those 2 files. I don't know...
## Issue Details MP_Float is not constructible from long. In itself, it wouldn't really matter, but Eigen uses long a lot, and this makes MP_Float and its derivatives (`Quotient`) unsuitable...
(based on a user report on cgal-discuss) ## Issue Details Writing a triangulation and reading it back produces an invalid triangulation. I've never used those functions before, but I think...
Hello, every time I run `apt upgrade`, I get a message > The currently running kernel version is 5.8.0-3-amd64 which is not the expected kernel version 5.9.0-1-amd64. I am aware...
Hello, currently, one can get a pair of filtration values, or a cocyle, but sometimes it is convenient to have access to something kind of intermediate: the 2 simplices that...
https://github.com/GUDHI/gudhi-devel/blob/71beeb391cef793836e2d91598f9f942748edbb0/src/Bitmap_cubical_complex/include/gudhi/Bitmap_cubical_complex_base.h#L254 We should add the corresponding typedefs directly so we don't need to derive from it.
For debugging purposes, it would be convenient in Python to be able to access the vector called "data" in CubicalComplex. It can be private, undocumented, have an ugly name starting...
In the documentation, we give 2 examples for TimeDelayEmbedding time-series = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] point cloud = [[1, 4, 7], [3, 6, 9]]...