Release v1.3.3

Release Notes

This is a minor release. The following changes were implemented:

• The 'derive3d' function now checks whether 'numbins' is an integer (as required by the function)
• The function responsible for estimating the area weighted grain size distribution (calc_areaweighted_grainsize) has been partially rewritten and optimized.

Fixed errors:

• The previous version of the script (v1.3.2) raised an error when trying to define a specific bin size using a number-weighted approach.

Planned features (v1.4):

• Estimate differential stress from grain size using different paleopizometric relations
• A new method to estimate an optimal kde bandwidth in multimodal populations.
• A new method to estimate a safe bin size (or number of classes) when using the Saltykov method.

Release v1.3.2

Release Notes

This is a minor release. The following changes were implemented:

• Now the plots are optimized for Matplotlib v2.0 or higher.
• The find_grain_size function now returns the interquartile range
• Improved functions output in the shell
• Minor change in the calc_area_weighted_grainsize function to prevent bad behavior when using Pandas data frame instead of Numpy arrays

Upcoming features (v1.4):

• New method to estimate an optimal kde bandwidth in multimodal populations.
• New method to estimate a safe bin size (or number of classes) when using the Saltykov method.

Release v1.3.1

Release Notes

This is a minor release. The following changes were implemented:

• Now the extract_areas function allows you to search and load the data through the file explorer of your operating system (i.e. you no longer have to define the file path manually). Also, it recognizes the type of file automatically.
• Several changes to prepare the script to the Matplotlib v2.0 new features.
• For consistency, the parameter 'addPerimeter' in the calc_diameters function was renamed as 'correct_diameter'.

Upcoming features:

• New method to estimate an optimal kde bandwidth in multimodal populations

Release v1.3

Release Notes

This is a major release. The following changes were implemented:

• Now the script requires Numpy v1.11 or higher to work properly. The script will check your Numpy version once you run it.
• The parameter type in the extract_areas function has been renamed as form to avoid the use of a Python keyword.
• Taking advantage of the new features implemented in Numpy v1.11, the find_grain_size function implements new methods to estimate the bin size in the histograms including an automatic mode (the default) and the Freedman-Diaconis', Sturge's, Scott's, Doane's, Rice's, and square root rules. See more details in the script documentation or here.
• The output showed in the shell after calling the find_grain_size function have been reorganized and improved for better clarity.
• The derive3D function automatically generates a text (csv) file containing the midpoints, class frequencies and cumulative volume estimated.
• The titles of the different plots have been renamed for better clarity.
• The get_Scott_binsize and get_FD_binsize functions have been deprecated.
• Improvements in the function specifications.
• Minor changes to meet the Style Guide for Python Code

Upcoming features:

• New method to estimate an optimal kde bandwidth in multimodal populations

Release v1.2

Release Notes

This is a major release. It adds new features and plots in the function find_grain_size.
For details see below:

• Now the user can define in the extract_areas function other column names different to the default name 'Areas'.
• The find_grain_size function allows estimating the logarithmic and square root grain sizes (including the generation of new plots).
• For better consistency, several functions responsible for generating the plots have been renamed.
• Minor cosmetic changes.

Upcoming features:

• New methods to estimate optimal bin sizes and bandwidths in multimodal populations

Release v1.1.1

Release Notes

This is a minor release.

• Now the same script is compatible with Python 2.7.x and 3.x versions.

Release v1.1

Release Notes

This is a major release. It fixes a known bug in the previous version.
For details see below:

• This release fixes a problem when using the two-step method in Python 3. Now, there are two different versions of the same source code, one for Python 2.7.x and one for Python 3.x versions. I am currently working to have a single compatible script (i.e. Py2-3) in the next version.
• The 'importdata' function has been renamed as 'extract_areas'.
• now it is indicated how to change the current directory when running the script.
• There is a general reorganization of the source code to facilitate future implementations, this includes: i) a simplification of the 'derive3D' function; ii) the addition of a new function named 'gen_xgrid'; iii) the relocation of the functions responsible for generating the plots.
• It is now possible to obtain the antispheres plot by changing a parameter in the 'Saltykov' function.
• A number of minor changes to improve the source code following the Style Guide for Python Code and other language corrections in the specifications of the functions

Release v1.0

Release Notes

This is a major release. The script contains a large number of changes and new features.
See below:

• GrainSizeTools script is now licensed under the Apache License, Version 2.0.
• The script now requires Pandas 0.16 or higher installed in the system.
• The importdata function now automatically extracts the data corresponding to the areas
  of grain profiles from the tabular-like data generated by the ImageJ application. It is
  no longer necessary to make intermediate steps between the ImageJ output and the GrainSizeTools
  script input.
• The organization of the script has changed significantly. All the code to generate the
  plots is now separated from the main functions. Specifically, each plot has their
  self-contained Python function. This facilitates to modify the parameters that control the
  aesthetics of the different plots.
• The GrainSizeTools script shows the Python current working directory when running the
  script for the first time.
• The Saltykov method implemented within the Saltykov function places the R parameter at the
  center of the classes (i.e. the midpoints) instead of at the upper edges of the bins (as in previous versions).
• Now the rule used by default to estimate the bin size in the find_grain_size function is
  the Scott rule instead of the Freedman-Diaconis rule.
• The derive3D function now includes several new features, including i) the ability to
  quantitatively estimate the volume occupied by a particular grain fraction defined by the
  user, ii) an experimental method to quantitatively estimating the actual 3D population of
  grain sizes when log-normal distributions are expected. See details in the documentation.
• A number of cosmetic changes either in the results and the plots.

For previous versions see: https://sourceforge.net/projects/grainsizetools/files/?source=navbar