Neutcurve

Latest version: v2.1.0

Safety actively analyzes 688512 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 1 of 6

2.1.0

-----

Added
+++++
- Added the ``draw_in_bounds`` option to the curve plotting so that curves never are plotted outside the data range. Addresses [this issue](https://github.com/jbloomlab/neutcurve/issues/59).

2.0.1

-----

Fixed
+++++
- Don't throw an uninterpretable error if ``CurveFits.fitParams`` called with no curves, instead just return empty data frame.

2.0.0

-----

Added
+++++
- The curve fitting parameters (top, bottom, slope) can now be constrained to a range in addition to being completely free or fixed. This can help with fitting some curves more sensibly (see [this issue](https://github.com/jbloomlab/neutcurve/issues/53)). Specifically:
- ``fixtop`` and ``fixbottom`` parameters to ``HillCurve`` can be 2-tuples of bounds
- added ``fixslope`` parameter to ``HillCurve`` and ``CurveFits``
- New ``constrain_params_range`` notebook tests and documents this functionality.

- Add ``no_curve_fit_first`` argument to ``HillCurve`` to aid debugging/development.

- Improvements to metrics for assessing curve fit (see [here](https://github.com/jbloomlab/neutcurve/issues/55#issuecomment-2016975219)):
- The coefficient of determination (``r2``) now is one if all points are fit by a straight line, rather than engative infinity.
- A root-mean-square-deviation (square root of mean residual) is now calculated as the ``rmsd`` attribute of ``HillCurve`` objects and reported in fit parameter summaries from ``CurveFits``.

1.1.2

-----

Fixed
+++++
- ``CurveFits.plotReplicates`` no longer fails if too many replicates, but instead recycles colors and markers. To make these combinations unique by default, also added another marker to ``CBMARKERS``.

1.1.1

-----

Fixed
+++++
- Fix internal bug in ``HillCurve`` that led to failure to try alternative fitting methods if initial fitting failed as it sometimes does for problematic curves.

1.1.0

-----

Added
+++++
- Report whether curve midpoint is in bounds of data (interpolated) or extrapolated, similar to as was already done for icXX:
- Added ``HillCurve.midpoint_bound`` and ``HillCurve.midpoint_bound_type`` attributes
- Report 'midpoint_bound' and 'midpoint_bound_type' in ``CurveFits.fitParams``

Page 1 of 6

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.