v0.8.11

Documentation: http://refl1d.readthedocs.org

2020-06-11 v0.8.11

• support list values for angle and wavelength in column data format
• use dQ from datafile rather than calculating it from dL and dT
• adjust dQ with sample broadening rather than calculating it from dL and dT
• fix json output so saving the fit doesn't raise an error
• fix functional magnetic profile so [rho, rho] is not treated as rho, theta.
• fix plots to show magnetic angle when not the standard magnetic angle
• fix legends when show magnetic angles on plots
• wxpython is now an optional dependency
• bumps: support limits=(min, max) for pm and pmp parameter ranges
• bumps: fix --entropy command line option
• bumps: parallel fitting suppressed in GUI for now---need to reuse thread pool

BREAKING CHANGE: default thetaM for functional magnetism changed from 0 to 270

See CHANGES.rst for complete history.

v0.8.10

Documentation: http://refl1d.readthedocs.org

2020-02-18 v0.8.10

• add --checkpoint=n to save mcmc state every n hours.
• hide resolution bars using Probe.show_resolution = False in your model.py
• fix residuals, fresnel, logfresnel, and q4 plots when interpolation is used
• fix error contour save function: rhoM was not being written
• fix simulation error: theory and simulated data were sharing the same vector
• fix doc generation: now builds with sphinx 2.4.0 and matplotlib 3.1.0.
• python 3.8 support requires bumps v0.7.14 (released 2020-01-03)

See CHANGES.rst for complete history.

v0.8.9

Documentation: http://refl1d.readthedocs.org

2019-10-15 v0.8.9

• json save: material and magnetism are now json objects
• json save: include non-parameter data in save file
• load4: accept 2 and 3 column data
• load4: override resolution and uncertainty given (or missing) from file
• load4: accept multi-entry data as Q probe without knowing theta/lambda
• load4: set the data slice to load
• load4: set default radiation to 'neutron'
• allow fit by number density for materials
• fix interpolation when plotting reflectivity between measured data points
• fix bug in dQ when sample_broadening is initialized to non-zeros
• revised installer: embedded python in zip file
• functions to compute transmission and reflection at each layer
• allow simulation with uncertainty from data
• force minimum uncertainty in data to 1e-11
• change default data view from Fresnel to log10
• apply resolution to saved Fresnel curve
• improved python 3 support

See CHANGES.rst for complete history.

v0.8.8

Documentation: http://refl1d.readthedocs.org

2019-03-01 v0.8.8

• fix json save for MixedExperiments
• save smooth magnetic profiles
• fix abeles code to choose correct branch cut below critical edge
• force absorption to be 0 or positive

See CHANGES.rst for complete history.

v0.8.8

Refl1d 0.8.8

v0.8.7

Documentation: http://refl1d.readthedocs.org

2018-12-18 v0.8.7

• make sample broadening a fittable parameter
• allow model + data to be loaded from zip file (bumps 0.7.12 and up)
• improve serialization support

See CHANGES.rst for complete history.

v0.8.6

Documentation: http://refl1d.readthedocs.org

2018-09-24 v0.8.6

• added serialization support
• added option to supply uncertainties when simulating data

See CHANGES.rst for complete history.

v0.8.5

Documentation: http://refl1d.readthedocs.org

2018-06-18 v0.8.5

• fix for plotting spin asymmetry when data is not present (model-only)
• added requirements to setup.py so that pip install refl1d suffices

2018-06-14 v0.8.4

• full support for python 3 in GUI
• allow :code:--pars=parfile with extra or missing parameters

2018-06-08 v0.8.3

• fix saved magnetic profiles

See CHANGES.rst for complete history.

v0.8.2

Documentation: http://refl1d.readthedocs.org

2018-05-18 v0.8.2

• include new entry points: run program by typing :code:refl1d at prompt

2018-05-17 v0.8.1

• allow alternate column order, such as :code:load4(..., columns="Q dQ R dR")
• include resolution effects in Fresnel reflectivity normalization
• allow magnetic profile to be saved

See CHANGES.rst for complete history.

v0.8.0

Documentation: http://refl1d.readthedocs.org

2017-12-01 v0.8.0

• incoherent cross sections now calculated as total minus coherent
• make sure displayed chisq is consistent with negative log likelihood
• allow blending across multiple interfaces
• allow Nevot-Croce calculations for magnetic models

See CHANGES.rst for complete history.