pjhaest / pest_tools Goto Github PK

PEST Tools 

Version 0.1.0 - Initial commit

Version 0.1.1 - added post processing of .res and/or .rei file.  
    Plot measured vs model, measured vs. residual, residual stats 

Version 0.1.2 - added histbin (2D histogram) plotting for residuals

Version 0.1.3 - added summary and plotting of contribution of each observation group to the objective function

Version 0.1.4 - added calculation of covariance and correlation matrix into pandas dataframe and associated plotting options

Description
-------------
Python modules used to aid in model calibration with PEST (Doherty, 2010).
 
Main goal is for quick development of visuals on important PEST output.

Current highlights include:
  - Read binary .jco file into pandas data frame
  - Calculate parameter sensitivity for all observations or with select observation groups removed
  - Calculate observation sensitivity
  - Quickly select and plot different views of parameter sensitivity (by group, most sensitive, least sensitive, etc.)
  - Read in output from JACTEST and plot data with interactive slider
  - Read in data from IDENTPAR and rank/plot
  - Read .rmr file from BEOPEST and plot a boxplot of run times by node
  - Read .res or .rei file and summarize.  Plot measured vs. residual, summarize contribution to objective function, residual statistics.
  - Calculate correlation matrix 
  - Plot "heat map" of correlation matrix
  - Plot dendrogram of correlation data
  - Plot "heat map" and dendrogram with smart sorting
  - Calculate covariance matrix
  - Calculate eigenvalues and eigen vectors

 
See examples for how things work (not complete).  

Dependencies
--------------
Listed are known to work, older versions may also work:
python 2.7.5
matplotlib 1.2.1
numpy 1.8
pandas 0.13.0

Installation
-------------
    Unzip files
    from directory of unzipped files run:
    $ python setup.py install

pest_tools will be installed the current python "site-packages" directory (e.g. C:\Python27\Lib\site-packages\pest_tools)

import in python using:
>>> import pest_tools 
In the examples and in practice PEST tools is commonly imported as follows:
>>> import pest_tools as pt

Reference
----------
Doherty, J., 2010, PEST, Model-independent parameter estimation—User 
manual, 5th ed.: Brisbane, Australia, Watermark Numerical Computing.