Eric O. LEBIGOT (EOL)

Results 48 issues of Eric O. LEBIGOT (EOL)

Could the install procedure from source (with `pip install ` instead of `python setup.py install`) be used for this project (the Python 2 compatibility may be an obstacle)? If yes,...

It makes sense, now, to have all the examples be valid Python 3 code. Print statements are the main change that should be applied.

Documentation

As of version https://github.com/lebigot/uncertainties/tree/59998e1760a8d3e18029fb1dde8d16e5757eaca9, the PyPI preparation script is broken: this is what continuous integration shows, as well as pip install in Python 2.7, when doing `import uncertainties`.

Sums are calculated in linear time, starting with version 3.0. But _cumulative_ sums are calculated in quadratic time (2 minutes for 10,000 terms, on my machine—much faster than even a...

Internals
New feature

See https://github.com/conda-forge/uncertainties-feedstock/issues/22

1. NumPy 1.17 introduced a **new argument** in `numpy.linalg.pinv()`. It would be better if uncertainties knew how to handle it (instead of refusing to use it). 2. NumPy 1.14 made...

NumPy+uncertainties

In order to catch more violations of the linear approximation hypothesis that underlies the calculations in the uncertainties module, it would make sense to have 1±NaN > 0 return False±Nan....

Behavior

From a user: > May I suggest you add an additional formatting specifier to "uncertainties"? > > In many cases, particularly in tabular presentation of data, one wants > to...

New feature

The documentation for `unumpy.isnan()` is the documentation for `numpy.vectorize()`. This is not helpful. It would be better to display the NumPy documentation, prefixed by a note about `unumpy.isnan()` being a...

Documentation
NumPy+uncertainties