changes in examples

README.md

@@ -4,24 +4,41 @@ Python connectivity module.
 It is a part of [GSOC 2015](http://www.google-melange.com/gsoc/project/details/google/gsoc2015/dokato/5649050225344512) project.
 Project blog is [here](http://dokato.github.io/connpy-blog/).
 
+## Content
+
+* Data - wrapper of your data
+* Connectivity - classes with connectivity estimators
+* Mvar fitting - this submodule includes some MVAR algorithms
+
+## License
+BSD 2-clause
+
+## Installation
+
 Documentation on [ReadTheDocs](http://connectivipy.readthedocs.org/).
 
-## Content
+## Installation
+
+Option 1: using GIT
+
+> git clone https://github.com/dokato/connectivipy.git
+> cd connectivipy
+> python setup.py install
 
-* Data 
-* Connectivity
-* Mvar fitting
+Option 2: Download ZIP from the menu on the right, unzip it and go
+in terminal to that folder. Than just execute:
+> python setup install
 
-##Changelog
+###Changelog
 
-### 0.34
+#### 0.34
 * short-time statistics
 * documentation in sphinx ready on readthedocs
 * visualization improved
 * conversion to trans3d
 * more examples
 
-### 0.31
+#### 0.31
 * connectivity methods: gDTF, gPDC, Coherency, PSI, GCI
 * new mvar estimation criterion: FPE
 * statistics for multitrial (bootstrap) and normal case (surrogate data)
@@ -30,15 +47,15 @@ Documentation on [ReadTheDocs](http://connectivipy.readthedocs.org/).
 * data plotting
 * working example
 
-### 0.2
+#### 0.2
 * connectivity methods: DTF, PDC, Partial Coherence, iPDC
 * mvar estimation criterions
 * mvar class static
 
-### 0.1
+#### 0.1
 * data class with simple preprocessing methods
 * mvar class almost done
 
-### 0.05
+#### 0.05
 * project structure
 * basic fitting

connectivipy/data.py

@@ -294,7 +294,6 @@ def plot_data(self, trial=0, show=True):
           *show* = True : boolean
             show the plot or not
         '''
-
         time = np.arange(0,self.__length)*1./self.__fs
         if self.__multitrial:
             plotdata = self.__data[:,:,trial]
@@ -437,13 +436,12 @@ def export_trans3d(self, filename='conn_trnas3d.dat', freq_band=[]):
           *freq_band* = [] : list
             frequency range [from_value, to_value] in Hz.            
         '''
-        trdef_str = [1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, -8, 0, 0, 0, 1]
-        tr_str = [0.0051, 0.0074, -3.4523, 0., -2.7277, 2.1161, 0.0005, 0., 2.1161, 2.7276, 0., -8., 0., 0., 0., 1.]
         content = ";electrodes = " + " ".join(self.__channames)
         content += "\r\n;start = -0.500000\r\n"
         content += ";samplerate = 12\r\n"
-        content += ";transform_default = " + " ".join([ str(x) for x in trdef_str[:self.__chans]]) + "\r\n"
-        content += ";transform = " + " ".join([ str(x) for x in tr_str[:self.__chans]]) + "\r\n"
+        # two following lines define initial position of head model
+        content += ";transform_default = 1 0 0 0 0 1 0 0 0 0 1 -8 0 0 0 1\r\n"
+        content += ";transform = 0.005148315144289768 0.007407087180879943 -3.4522594293647604 0.0 -2.727691946877633 2.116098522619611 0.000472475639 0.0 2.1160923126171305 2.7276819307525173 0.009008154977586572 -8.0 0.000000 0.000000 0.000000 1.000000\r\n"
         content += "\r\n"
         # integrate value of estimator in given frequency band
         freqs = np.linspace(0, int(self.__fs/2), self.__estim.shape[0])
@@ -460,7 +458,6 @@ def export_trans3d(self, filename='conn_trnas3d.dat', freq_band=[]):
             fl.write(content)
 
     # auxiliary methods:
-
     def __make_params_dict(self, args):
         """
         Making list of parameters from *self._parameters*

connectivipy/mvarmodel.py

@@ -1,5 +1,4 @@
 # -*- coding: utf-8 -*-
-# -*- coding: utf-8 -*-
 #! /usr/bin/env python
 
 import numpy as np

doc/tutorial.rst

@@ -1,9 +1,18 @@
 .. _tutorial:
 
-Tutorial
+Tutorials
 ==================
 
-All examples and tutorial will go here.
+Installation
+########
+
+The easiest way is to use *GIT* and just execute:
+
+.. code-block:: bash
+
+    $ git clone https://github.com/dokato/connectivipy.git
+    $ cd connectivipy
+    $ python setup.py install
 
 Data class example
 ########
@@ -14,10 +23,14 @@ Data class example
 
     import numpy as np 
     import connectivipy as cp
-    from connectivipy.mvar.fitting import mvar_gen
+    from connectivipy import mvar_gen
 
-    # MVAR model coefficients
+    ### MVAR model coefficients
+
+    # let's build mvar model matrix
     A = np.zeros((2, 5, 5))
+    # 2 - first dimension is model order
+    # 5 - second and third dimensions mean number of channels
     A[0, 0, 0] = 0.95 * 2**0.5
     A[1, 0, 0] = -0.9025
     A[0, 1, 0] = -0.5
@@ -28,36 +41,39 @@ Data class example
     A[0, 4, 3] = -0.25 * 2**0.5
     A[0, 4, 4] = 0.25 * 2**0.5
 
-    # multitrial signal generation
-    ysig = np.zeros((5,10**3,5))
+    # multitrial signal generation from a matrix above
+    # let's generate 5-channel signal with 1000 data points
+    # and 5 trials using function mvar_gen
+    ysig = np.zeros((5, 10**3, 5))
     ysig[:,:,0] = mvar_gen(A,10**3)
     ysig[:,:,1] = mvar_gen(A,10**3)
     ysig[:,:,2] = mvar_gen(A,10**3)
     ysig[:,:,3] = mvar_gen(A,10**3)
     ysig[:,:,4] = mvar_gen(A,10**3)
 
-    # connectivity analysis
+    #### connectivity analysis
+
     data = cp.Data(ysig,128, ["Fp1", "Fp2","Cz", "O1","O2"])
 
-    # you may want to plot data (in multitrial case average along trials
-    # is shown)
+    # you may want to plot data (in multitrial case only one trial is shown)
     data.plot_data()
 
-    # fit mvar using specific algorithm
+    # fit mvar using Yule-Walker algorithm and order 2
     data.fit_mvar(2,'yw')
 
     # you can capture fitted parameters and residual matrix
     ar, vr = data.mvar_coefficients 
 
-    # connectivity estimators
+    # now we investigate connectivity using gDTF
     gdtf_values = data.conn('gdtf')
-    #gdtf_significance = data.significance(Nrep=200, alpha=0.05)
+    gdtf_significance = data.significance(Nrep=200, alpha=0.05)
     data.plot_conn('gDTF')
 
     # short time version with default parameters
-    pdc_shorttime = data.short_time_conn('pdc', nfft=1, no=10)
+    pdc_shorttime = data.short_time_conn('pdc', nfft=100, no=10)
     data.plot_short_time_conn("PDC")
 
+
 How to use specific classes
 ########
 

examples/example1.py

@@ -3,10 +3,14 @@
 
 import numpy as np 
 import connectivipy as cp
-from connectivipy.mvar.fitting import mvar_gen
+from connectivipy import mvar_gen
 
-# MVAR model coefficients
+### MVAR model coefficients
+
+# let's build mvar model matrix
 A = np.zeros((2, 5, 5))
+# 2 - first dimension is model order
+# 5 - second and third dimensions mean number of channels
 A[0, 0, 0] = 0.95 * 2**0.5
 A[1, 0, 0] = -0.9025
 A[0, 1, 0] = -0.5
@@ -17,30 +21,32 @@
 A[0, 4, 3] = -0.25 * 2**0.5
 A[0, 4, 4] = 0.25 * 2**0.5
 
-# multitrial signal generation
-ysig = np.zeros((5,10**3,5))
+# multitrial signal generation from a matrix above
+# let's generate 5-channel signal with 1000 data points
+# and 5 trials using function mvar_gen
+ysig = np.zeros((5, 10**3, 5))
 ysig[:,:,0] = mvar_gen(A,10**3)
 ysig[:,:,1] = mvar_gen(A,10**3)
 ysig[:,:,2] = mvar_gen(A,10**3)
 ysig[:,:,3] = mvar_gen(A,10**3)
 ysig[:,:,4] = mvar_gen(A,10**3)
 
-# connectivity analysis
+#### connectivity analysis
+
 data = cp.Data(ysig,128, ["Fp1", "Fp2","Cz", "O1","O2"])
 
-# you may want to plot data (in multitrial case average along trials
-# is shown)
+# you may want to plot data (in multitrial case only one trial is shown)
 data.plot_data()
 
-# fit mvar using specific algorithm
+# fit mvar using Yule-Walker algorithm and order 2
 data.fit_mvar(2,'yw')
 
 # you can capture fitted parameters and residual matrix
 ar, vr = data.mvar_coefficients 
 
-# connectivity estimators
+# now we investigate connectivity using gDTF
 gdtf_values = data.conn('gdtf')
-#gdtf_significance = data.significance(Nrep=200, alpha=0.05)
+gdtf_significance = data.significance(Nrep=200, alpha=0.05)
 data.plot_conn('gDTF')
 
 # short time version with default parameters

examples/example2.py

@@ -1,11 +1,12 @@
 import numpy as np
 import matplotlib.pyplot as plt
 import connectivipy as cp
-from connectivipy.mvar.fitting import mvar_gen
+from connectivipy import mvar_gen
 
 fs = 256.
 acf = np.zeros((3,3,3))
-# matrix shape meaning (p,k,k) k - number of channels,
+# matrix shape meaning 
+# (p,k,k) k - number of channels,
 # p - order of mvar parameters
 
 acf[0,0,0] = 0.3
@@ -22,7 +23,7 @@
 # assign static class cp.Mvar to variable mv
 mv = cp.Mvar
 
-# find best model order
+# find best model order using Vieira-Morf algorithm
 best, crit = mv.order_akaike(y,15,'vm')
 plt.plot(1+np.arange(len(crit)),crit,'g')
 plt.show()