This content refers to an unreleased development version of PyMVPA

To provide the most recent news and documentation www.pymvpa.org reflects the development 0.6 series of PyMVPA. If you are interested in the documentation of the previous stable 0.4 series of PyMVPA, please visit v04.pymvpa.org.

PDF version of the PyMVPA Manual

The PDF version of the manual is available for download.

• Introduction
  • What this Manual is NOT
  • A bit of History
  • Authors & Contributors
  • How to cite PyMVPA
    • Peer-reviewed publications
    • Posters
  • Acknowledgements
• Installation
  • Dependencies
    • Must Have
    • Strong Recommendations
    • Suggestions
  • Installing Binary Packages
    • Debian
    • Debian backports and inofficial Ubuntu packages
    • Windows
    • MacOS X
  • Building from Source
    • Three Ways to Obtain the Sources
    • Build it (General instructions)
    • Build with enabled LIBSVM bindings
    • Alternative build procedure
    • Windows
    • OpenSUSE
    • Fedora
    • MacOS X
• Getting Started
  • For the Impatient
  • Module Overview
• Tutorial Introduction to PyMVPA
  • Tutorial Prerequisites
    • What Do I Need To Get Python Running
    • Recommended Reading and Viewing
  • Part 1: A Gentle Start
    • Getting the data
    • Dealing With A Classifier
    • Cross-validation
    • References
  • Part 2: Dataset Basics and Concepts
    • Attributes
    • Slicing, resampling, feature selection
    • Loading fMRI data
    • Storage
  • Part 3: Mappers – The Swiss Army Knife
    • Doing get_haxby2001_data() From Scratch
    • There and back again – a Mapper’s tale
  • Part 4: Classifiers – All Alike, Yet Different
    • We Need To Take A Closer Look
    • Meta-Classifiers To Make Complex Stuff Simple
  • Part 5: Searchlite
    • Measures
    • Searching, searching, searching, ...
    • For real!
  • Part 6: Looking Without Searching – Sensitivity Analysis
    • It’s A Kind Of Magic
    • Thanks For The Fish
    • Dissect The Classifier
    • Closing Words
  • Part 7: “When Is The Signal” – Event-related Data Analysis
    • Event-related Pre-processing Is Not Event-related
    • From Timeseries To Events
    • A Plotting Example
    • If That Was Easy...
    • After all is done
  • Part 8: The Earth Is Round – Significance Testing
    • Statistical Tools in Python
    • Dataset Exploration for Confounds
    • Hypothesis Testing
    • Statistical Treatment of Sensitivities
    • References
• Miscellaneous
  • Managing (Custom) Configurations
  • Progress Tracking
    • Redirecting Output
    • Verbose Messages
    • Warning Messages
    • Debug Messages
    • PyMVPA Status Summary
  • Additional Little Helpers
    • Random Number Generation
    • Unittests at a Grasp
  • FSL Bindings
• Example Analyses and Scripts
  • Preprocessing
    • Visualization of Data Projection Methods
    • Simple Data-Exploration
  • Analysis
    • Separating hyperplane tutorial
    • Tiny Example of a Full Cross-Validation
    • Compare SMLR to Linear SVM Classifier
    • Classifier Sweep
    • The effect of different hyperparameters in GPR
    • Simple model selection: grid search for GPR
    • Efficient cross-validation using a cached kernel
    • Minimal Searchlight Example
    • Searchlight on fMRI data
    • A searchlight computing a dissimilarity matrix measure
    • Sensitivity Measure
    • Classification of SVD-mapped Datasets
    • Monte-Carlo testing of Classifier-based Analyses
    • Nested Cross-Validation
    • Determine the Distribution of some Variable
    • Spatio-temporal Analysis of event-related fMRI data
    • MNIST handwritten digits
    • Analysis of eye movement patterns
  • Visualization
    • ERP/ERF-Plots
    • kNN – Model Flexibility in Pictures
    • Simple Plotting of Classifier Behavior
    • Generating Topography plots
    • Self-organizing Maps
    • Basic (f)MRI plotting
  • Miscellaneous
    • Kernel-Demo
    • Curve-Fitting
• Frequently Asked Questions
  • General
    • I’m a Matlab user. How hard is learning Python and PyMVPA for me?
    • It is sloooooow. What can I do?
    • I am tired of writing these endless import blocks. Any alternative?
    • I feel like I want to contribute something, do you mind?
    • I want to develop a new feature for PyMVPA. How can I do it efficiently?
    • The manual is quite insufficient. When will you improve it?
  • Data import, export and storage
    • What file formats are understood by PyMVPA?
    • What if there is no special file format for some particular datatype?
  • Data preprocessing
    • Is there an easy way to remove invariant features from a dataset?
    • How can I do block-averaging of my block-design fMRI dataset?
  • Data analysis
    • How do I know which features were finally selected by a classifier doing feature selection?
    • How do I extract sensitivities from a classifier used within a cross-validation?
    • Can PyMVPA deal with literal class labels?
• Glossary
• References
• License
  • 3rd Party Code
    • LIBSVM
    • NIPY
    • PDFBook
• Development Changelog
  • Releases