-
-
Notifications
You must be signed in to change notification settings - Fork 9
/
Copy pathCHANGELOG
52 lines (52 loc) · 1.69 KB
/
CHANGELOG
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
0.4.1
- docs: remove Python 2 reference
0.4.0
- ref: support NumPy 2
- setup: migrate to GitHub Actions
- setup: migrate to pyproject.toml
- docs: remove dependency on old mock standalone library (#19)
- docs: overhaul and add changelog
0.3.3
- fix: add constant `ZERO_CUTOFF` that defines when a trace average is
treated as zero
- enh: replace asserts with raises (#15)
0.3.2
- fix: multipletau package not available when building the docs
0.3.1
- fix: detect correct version when building the docs from source tar ball
0.3.0
- feat: add option to choose the strategy for propagating values to
the next register (#14)
- feat: add option to return the pure sum and the internal normalization
count (#14)
0.2.0
- tests: filter warnings and check with flake8
- implement unique warning classes
0.1.9
- include docs in sdist
0.1.8
- update documentation and example files
- move documentation to readthedocs.io
0.1.7
- code cleanup with pep8 and autopep8
- always use numpy dtypes
- fix tests:
- take into account floating inaccuracies
- support i386 numpy dtypes
0.1.6
- also compute correlation for zero lag time (`G(tau==0)`)
- support NumPy 1.11
- add tests to complete code coverage
- bugfixes:
- wrong normalization for cplx array `v` in `correlate` if `normalize==True`
- wrong normalization in `correlate_numpy` if `normalize==False`
0.1.5
- update documentation
- support Python 3
0.1.4
- integer and boolean input types are now automatically converted to floats
- `multipletau.correlate` now works with complex data types
- `multipletau.correlate` now checks if input data are same objects
- documentation now contains examples
0.1.3
- first non-cython implementation