heat
heat copied to clipboard
Distributed tensors and Machine Learning framework with GPU and MPI acceleration in Python
### What happened? `axis=None` in `any`/`all` performs a reduction across all dimensions of the array. However, `keepdim=True` does not preserve shape in this case. ### Code snippet triggering the error...
### What happened? Loss of precision is observed in several Heat functions when using the `float64` data type. This is due to pre-conversion to `float32` (the pytorch default floating data...
**Feature functionality** Enable the `copy` kwarg in `asarray`. **Additional context** In compliance with the Python array API standard
## Description # DRAFT ## Intro: main points from Numpy array indexing scheme ## Heat: indexing massive, memory-distributed arrays ### process-local indexing ### distributed indexing - sorted key - non-sorted...
## Description A first idea of the cholesky decomposition. It does not check if the input is a Hermitian, positive-definite matrix. *It doesn't work for distributed matrices with complex values.*...
## Description Issue/s resolved: #980 ## Changes proposed: - Wait for precommit - - - ## Type of change ## Memory requirements ## Performance ## Due Diligence - [ ]...
## Description jitting for skew and kurtosis. possibly more changes to come here, this is the branch to be used during benchmarking and testing of the moment functions ## Changes...
Issue/s resolved: #351 ## Type of change Documentation update ## Due Diligence - [ ] All split configurations tested - [ ] Multiple dtypes tested in relevant functions - [...
## Description Replaced factories.array with DNDarray in `io.py` Issue/s resolved: https://github.com/helmholtz-analytics/heat/issues/797 ## Changes proposed: - factories.array -> dndarray.DNDarray ## Type of change enhancement ## Due Diligence - [ ] All...
## Description implement the `__partitioned__` attribute to the DNDarray for compatibility with daal4py (https://github.com/IntelPython/DPPY-Spec/issues/3). At the moment, this is not used by heat internally. However, There are some ideas about...