flow_stability

flow_stability copied to clipboard

The dict of clusterings should be created during the initialization so that they can be passed to the parallel clustering. Alternatively, it should be possible to run the parallel clustering...

alexbovet

here: https://github.com/alexbovet/flow_stability/blob/main/parallel_clustering.py

alexbovet

It's not clear how to use it and where the clusterings are stored. A small example should be added to the doc. https://github.com/alexbovet/flow_stability/blob/9d187a1d85eb396275ac6f9b6ad7dd1e99cfa3fd/FlowStability.py#L1679

alexbovet

Starting semver versioning

3

This implements: - CHANGELOG.md with automated updates - Workflows to prepare and release new version - Semver based versioning

j-i-l

Write tests for core functions

11

j-i-l

First draft of Summary

1

> A summary describing the high-level functionality and purpose of the software for a diverse, non-specialist audience. _[source](https://joss.readthedocs.io/en/latest/paper.html)_ To work on this issue, commit to this branch: https://github.com/alexbovet/flow_stability/tree/25-first-draft-of-summary 1. Checkout...

j-i-l

Sparse autocov matrix classes

3

Currently we have two of them: https://github.com/alexbovet/flow_stability/blob/8648a8bd715e3b086ed13cb7453d41754c176398/src/flowstab/SparseStochMat.py#L1175-L1176 and https://github.com/alexbovet/flow_stability/blob/8648a8bd715e3b086ed13cb7453d41754c176398/src/flowstab/SparseStochMat.py#L1454-L1455 With a much larger usage pattern for `sparse_autocov_mat`. Actually, `sparse_autocov_csr_mat` only occurs twice as potential substitute for `sparse_autocov_mat' [here](https://github.com/alexbovet/flow_stability/blob/8648a8bd715e3b086ed13cb7453d41754c176398/src/flowstab/FlowStability.py#L1443-L1444) and [here](https://github.com/alexbovet/flow_stability/blob/8648a8bd715e3b086ed13cb7453d41754c176398/src/flowstab/FlowStability.py#L1397-L1398)...

j-i-l

Renaming in SparseStochMat module and others

2

The convention is (`scipy` is actually a terrible example for this): - `snake_case` for module names - `PascalCase` for class names so: SparseStochMat.py -> sparse_stoch_mat.py `class sparse_stoch_mat` -> `class SparseStochMat`...

j-i-l

Index sorting in the `csr_` operations

5

> Comparing the output of the [`csr_`](https://github.com/alexbovet/flow_stability/blob/ee5e48f3e75eabec91389d90e3cf1cb60e25f282/src/flowstab/_cython_sparse_stoch.pyx) methods with native implementations we get mismatches on `.data` and `.indices` (see e.g. https://github.com/alexbovet/flow_stability/actions/runs/10865864561/job/30152834847#step:6:102). > > A smaller example illustrates this better (also...

j-i-l

Type declaration for `indices` and `indptr` might lead to dtype mismatch

3

When passing the arguments of a `scipy.sparse.csr_matrix` to `sparse_stoch_from_full_csr` we run into a `ValueError`: >>> > long long[:] Tf_indices, E ValueError: Buffer dtype mismatch, expected 'long long' but got 'int'...

j-i-l