heat

heat copied to clipboard

`ht.histogram()` numpy API compliance

2

The current ht.histogram() function calls the ht.histc() function which uses PyTorch. It should work more like NumPy instead. _Originally posted by @mtar in https://github.com/helmholtz-analytics/heat/issues/547#issuecomment-717151678_

ClaudiaComito

ht.average() should allow weighted average across multiple axes

2

**Feature functionality** `ht.average(x, weights=None, axis=None, returned=False)` (PR #352) has the same constraints as numpy.average(), among them the following: - if axis is not None, the weights tensor must be 1D...

ClaudiaComito

Implement AD functionality for MPI operations

2

**Feature functionality** It is planned to implement a thin provisiong layer above the existing MPI communication API that allows for nearly seemless integration of the MPI calls with the PyTorch...

d1saster

Implement isin() (and in1d())

3

**Related** #340 **Feature functionality** https://docs.scipy.org/doc/numpy/reference/generated/numpy.in1d.html https://docs.scipy.org/doc/numpy/reference/generated/numpy.isin.html

ClaudiaComito

Implement SVD

1

**Related** #343 QR, possibly #336, testing -> #390 **Feature functionality** Calculate the singular value decomposition of a matrix

coquelin77

Support `savez`/`savez_compressed` and loading NpzFiles

1

Any chance for supporting `savez`/`savez_compressed` and loading NpzFiles? _Originally posted by @fschlimb in https://github.com/helmholtz-analytics/heat/issues/101#issuecomment-843967591_ References: - https://numpy.org/doc/stable/reference/generated/numpy.savez.html - https://numpy.org/doc/stable/reference/generated/numpy.savez_compressed.html

ClaudiaComito

Clean up concatenate using redistribute

2

**Feature functionality** redistribute can be used to clean up the concatenate function, reduce the complexity of the code, and clarify what it does to a 3rd party.

coquelin77

Set up GitHub Actions

3

**Feature functionality** GitHub introduced the GitHub Actions. This is a CI/CD pipeline similar to the one GitLab already provides. The benefit is that each build can be split into smaller...

TheSlimvReal

Provide a matrix gallery for testing

1

**Related** This is mostly related to the current LA efforts. **Feature functionality** We need some functions that can generate matrices of arbitrary size that have certain unique features that allow...

d1saster

Same numbers for random number on different split axis

6

**Description** Currently the random array returned by `ht.random.rand` is different depending on the split of the tensor. For example: ``` >>> ht.random.seed(1) >>> a = ht.random.randn(4, 4, split=0) >>> ht.random.seed(1)...

TheSlimvReal