Skip to content

Commit

Permalink
updated
Browse files Browse the repository at this point in the history
  • Loading branch information
Jakob Runge committed May 23, 2017
1 parent 136e821 commit 95c5ac9
Show file tree
Hide file tree
Showing 24 changed files with 1,930 additions and 563 deletions.
9 changes: 4 additions & 5 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -10,8 +10,8 @@ Version 3.0 described in http://arxiv.org/abs/1702.07007

## General Notes

TIGRAMITE is a time series analysis python module. It allows to reconstruct graphical models (conditional independence graphs) from discrete or continuously-valued time series based on a causal discovery algorithm and create high-quality plots of the results.

TIGRAMITE is a time series analysis python module. It implements the PCMCI causal discobery method that allows to reconstruct graphical models (conditional independence graphs) from discrete or continuously-valued time series and includes modules for high-quality plots of the results.
TODO: also mediation, prediction

## Features

Expand All @@ -24,6 +24,7 @@ TIGRAMITE is a time series analysis python module. It allows to reconstruct grap
- handling of missing values and masks
- p-value correction and confidence interval estimation
- can be used for feature selection and prediction
TODO: also mediation, prediction


## Required python packages
Expand Down Expand Up @@ -70,9 +71,7 @@ You commit to cite TIGRAMITE in your reports or publications if used:

## License

Copyright (C) 2012-2017 Jakob Runge

mpi4py wrapper module "mpi.py" Copyright (C) 2012 Jobst Heitzig
Copyright (C) Jakob Runge

See license.txt for full text.

Expand Down
Binary file modified docs/_build/doctrees/environment.pickle
View file
Binary file not shown.
Binary file modified docs/_build/doctrees/index.doctree
View file
Binary file not shown.
11 changes: 10 additions & 1 deletion docs/_build/html/_modules/tigramite/data_processing.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -93,12 +93,20 @@ <h1>Source code for tigramite.data_processing</h1><div class="highlight"><pre>
<span class="sd"> </span>
<span class="sd"> mask : array-like, optional (default: None)</span>
<span class="sd"> Optional mask array, must be of same shape as data</span>

<span class="sd"> missing_flag : number, optional (default: None)</span>
<span class="sd"> Flag for missing values in dataframe. Dismisses all time slices of</span>
<span class="sd"> samples where missing values occur in any variable and also flags</span>
<span class="sd"> samples for all lags up to 2*tau_max. This avoids biases, see</span>
<span class="sd"> section on masking in Supplement of [1]_.</span>

<span class="sd"> &quot;&quot;&quot;</span>
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data</span><span class="p">,</span>
<span class="n">mask</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
<span class="n">mask</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span> <span class="n">missing_flag</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>

<span class="bp">self</span><span class="o">.</span><span class="n">values</span> <span class="o">=</span> <span class="n">data</span>
<span class="bp">self</span><span class="o">.</span><span class="n">mask</span> <span class="o">=</span> <span class="n">mask</span>
<span class="bp">self</span><span class="o">.</span><span class="n">missing_flag</span> <span class="o">=</span> <span class="n">missing_flag</span>
<span class="n">T</span><span class="p">,</span> <span class="n">N</span> <span class="o">=</span> <span class="n">data</span><span class="o">.</span><span class="n">shape</span>

<span class="k">if</span> <span class="nb">type</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">values</span><span class="p">)</span> <span class="o">!=</span> <span class="n">numpy</span><span class="o">.</span><span class="n">ndarray</span><span class="p">:</span>
Expand All @@ -115,6 +123,7 @@ <h1>Source code for tigramite.data_processing</h1><div class="highlight"><pre>
<span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;Mask array must of same shape as data array&quot;</span><span class="p">)</span></div>



<div class="viewcode-block" id="lowhighpass_filter"><a class="viewcode-back" href="../../index.html#tigramite.data_processing.lowhighpass_filter">[docs]</a><span class="k">def</span> <span class="nf">lowhighpass_filter</span><span class="p">(</span><span class="n">data</span><span class="p">,</span> <span class="n">cutperiod</span><span class="p">,</span> <span class="n">pass_periods</span><span class="o">=</span><span class="s1">&#39;low&#39;</span><span class="p">):</span>
<span class="sd">&quot;&quot;&quot;Butterworth low- or high pass filter.</span>

Expand Down
44 changes: 19 additions & 25 deletions docs/_build/html/_modules/tigramite/independence_tests.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -122,11 +122,11 @@ <h1>Source code for tigramite.independence_tests</h1><div class="highlight"><pre
<span class="sd"> use_mask : bool, optional (default: False)</span>
<span class="sd"> Whether a supplied mask should be used.</span>

<span class="sd"> mask_type : {&#39;missing&#39;,&#39;y&#39;,&#39;x&#39;,&#39;z&#39;,&#39;xy&#39;,&#39;xz&#39;,&#39;yz&#39;,&#39;xyz&#39;}</span>
<span class="sd"> Masking mode: Either &#39;missing&#39; or indicators for which variables in the</span>
<span class="sd"> dependence measure I(X; Y | Z) the samples should be masked. If None,</span>
<span class="sd"> &#39;missing&#39; is used, which excludes masked samples for all X, Y, and Z up</span>
<span class="sd"> to tau_max if a value is missing in any of them. Explained in [1]_.</span>
<span class="sd"> mask_type : {&#39;y&#39;,&#39;x&#39;,&#39;z&#39;,&#39;xy&#39;,&#39;xz&#39;,&#39;yz&#39;,&#39;xyz&#39;}</span>
<span class="sd"> Masking mode: Indicators for which variables in the dependence measure</span>
<span class="sd"> I(X; Y | Z) the samples should be masked. If None, &#39;y&#39; is used, which</span>
<span class="sd"> excludes all time slices containing masked samples in Y. Explained in</span>
<span class="sd"> [1]_.</span>

<span class="sd"> significance : str, optional (default: &#39;analytic&#39;)</span>
<span class="sd"> Type of significance test to use. In this package &#39;analytic&#39;, </span>
Expand Down Expand Up @@ -206,7 +206,7 @@ <h1>Source code for tigramite.independence_tests</h1><div class="highlight"><pre
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_mask</span><span class="p">:</span>
<span class="bp">self</span><span class="o">.</span><span class="n">recycle_residuals</span> <span class="o">=</span> <span class="kc">False</span>
<span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">mask_type</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
<span class="bp">self</span><span class="o">.</span><span class="n">mask_type</span> <span class="o">=</span> <span class="s1">&#39;missing&#39;</span>
<span class="bp">self</span><span class="o">.</span><span class="n">mask_type</span> <span class="o">=</span> <span class="s1">&#39;y&#39;</span>

<span class="c1"># if use_mask:</span>
<span class="c1"># if mask_type is None or len(set(mask_type) -</span>
Expand All @@ -215,27 +215,21 @@ <h1>Source code for tigramite.independence_tests</h1><div class="highlight"><pre
<span class="c1"># % mask_type + &quot; &#39;x&#39;,&#39;y&#39;,&#39;z&#39;, or any &quot;</span>
<span class="c1"># &quot;combination&quot;)</span>

<div class="viewcode-block" id="CondIndTest.set_dataframe"><a class="viewcode-back" href="../../index.html#tigramite.independence_tests.CondIndTest.set_dataframe">[docs]</a> <span class="k">def</span> <span class="nf">set_dataframe</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataframe</span><span class="p">,</span> <span class="n">missing_flag</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
<div class="viewcode-block" id="CondIndTest.set_dataframe"><a class="viewcode-back" href="../../index.html#tigramite.independence_tests.CondIndTest.set_dataframe">[docs]</a> <span class="k">def</span> <span class="nf">set_dataframe</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">dataframe</span><span class="p">):</span>
<span class="sd">&quot;&quot;&quot;Initialize dataframe.</span>

<span class="sd"> Parameters</span>
<span class="sd"> ----------</span>
<span class="sd"> dataframe : data object</span>
<span class="sd"> This can either be the tigramite dataframe object or a pandas data</span>
<span class="sd"> frame. It must have the attributes dataframe.values yielding a numpy</span>
<span class="sd"> array of shape (observations T, variables N) and optionally a mask</span>
<span class="sd"> of the same shape. </span>

<span class="sd"> missing_flag : number, optional (default: None)</span>
<span class="sd"> Flag for missing values in dataframe. Dismisses all time slices of</span>
<span class="sd"> samples where missing values occur in any variable and also flags</span>
<span class="sd"> samples for all lags up to 2*tau_max. This avoids biases, see</span>
<span class="sd"> section on masking in Supplement of [1]_.</span>
<span class="sd"> Set tigramite dataframe object. It must have the attributes</span>
<span class="sd"> dataframe.values yielding a numpy array of shape (observations T,</span>
<span class="sd"> variables N) and optionally a mask of the same shape and a missing</span>
<span class="sd"> values flag.</span>

<span class="sd"> &quot;&quot;&quot;</span>
<span class="bp">self</span><span class="o">.</span><span class="n">data</span> <span class="o">=</span> <span class="n">dataframe</span><span class="o">.</span><span class="n">values</span>
<span class="bp">self</span><span class="o">.</span><span class="n">mask</span> <span class="o">=</span> <span class="n">dataframe</span><span class="o">.</span><span class="n">mask</span>
<span class="bp">self</span><span class="o">.</span><span class="n">missing_flag</span> <span class="o">=</span> <span class="n">missing_flag</span></div>
<span class="bp">self</span><span class="o">.</span><span class="n">missing_flag</span> <span class="o">=</span> <span class="n">dataframe</span><span class="o">.</span><span class="n">missing_flag</span></div>

<span class="k">def</span> <span class="nf">_keyfy</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="n">z</span><span class="p">):</span>
<span class="sd">&quot;&quot;&quot;Helper function to make lists unique.&quot;&quot;&quot;</span>
Expand Down Expand Up @@ -2082,13 +2076,13 @@ <h1>Source code for tigramite.independence_tests</h1><div class="highlight"><pre
<div class="viewcode-block" id="CMIknn"><a class="viewcode-back" href="../../index.html#tigramite.independence_tests.CMIknn">[docs]</a><span class="k">class</span> <span class="nc">CMIknn</span><span class="p">(</span><span class="n">CondIndTest</span><span class="p">):</span>
<span class="sd">r&quot;&quot;&quot;Conditional mutual information test based on nearest-neighbor estimator.</span>

<span class="sd"> Conditional mutual information is the most general dependency measure</span>
<span class="sd"> coming from an information-theoretic framework. It makes no assumptions</span>
<span class="sd"> about the parametric form of the dependencies by directly estimating the</span>
<span class="sd"> underlying joint density. The test here is based on the estimator in S.</span>
<span class="sd"> Frenzel and B. Pompe, Phys. Rev. Lett. 99, 204101 (2007), combined with a</span>
<span class="sd"> shuffle test to generate the distribution under the null hypothesis of</span>
<span class="sd"> independence. The knn-estimator is suitable only for variables taking a </span>
<span class="sd"> Conditional mutual information is the most general dependency measure coming</span>
<span class="sd"> from an information-theoretic framework. It makes no assumptions about the</span>
<span class="sd"> parametric form of the dependencies by directly estimating the underlying</span>
<span class="sd"> joint density. The test here is based on the estimator in S. Frenzel and B.</span>
<span class="sd"> Pompe, Phys. Rev. Lett. 99, 204101 (2007), combined with a shuffle test to</span>
<span class="sd"> generate the distribution under the null hypothesis of independence first</span>
<span class="sd"> used in [3]_. The knn-estimator is suitable only for variables taking a</span>
<span class="sd"> continuous range of values. For discrete variables use the CMIsymb class.</span>

<span class="sd"> Notes</span>
Expand Down
Loading

0 comments on commit 95c5ac9

Please sign in to comment.