MDF Data Flow Toolkit

MDF is a Python package that aims to make writing code simpler, particularly for code that is intended to be run multiple times with varying inputs.

• Introduction
• Nodes
  • Time Dependent Nodes
    • Filtering
  • Queue Nodes
  • Other Node Types
    • Delay Node
    • NaN Sum Node
    • Cumulative Product Node
    • Apply Node
  • Method Syntax For Node Types
• Class Nodes
• Contexts
  • Setting Values
  • Overriding Nodes
  • Shifted Contexts
• Back-Testing and Scenario Analysis
  • Back-Testing
    • build_dataframe
    • plot
    • to_csv
    • get_final_values
    • run
  • Scenario Analysis
    • scenario
    • plot_surface
    • heatmap
• Interactive use of MDF in IPython
  • Working with Timeseries
  • Working with Data
  • Applying Functions
  • Accessing the Context and Shifting
  • Using the MDF Viewer
• API Reference
  • Nodes Types
  • Node Factory Functions
  • Custom Node Types
  • Pre-defined Nodes
  • Functions
  • Classes
    • MDFContext
    • MDFNode
    • MDFEvalNode
    • CSVWriter
    • DataFrameBuilder
    • FinalValueCollector
• Implementation Details
  • Building MDF
    • Use of Cython
    • Compiling MDF
    • Running/Debugging Non-Cythoned MDF
    • Debugging the Compiled Cython Code
    • Profiling MDF
  • Source Code Overview
    • context.py
    • nodes.py
    • cqueue.py
    • nodetypes.py
    • runner.py
    • to_dot.py
    • parser.py
    • ctx_pickle.py
    • builders sub-package
    • io sub-package
    • pylab sub-package
    • regression sub-package
    • remote sub-package
    • tests sub-package
  • Node Evaluation
    • Dependencies
    • Node evaluation
    • Shifted contexts
  • Node Types
    • Generators and iterators
    • Method syntax for node types
  • Class nodes

Indices and tables

• Index
• Module Index
• Search Page