new possibility of plotting and new example of usage

connectivipy/__init__.py

@@ -11,7 +11,8 @@
 # plain plotting from values
 def plot_conn(values, name='', fs=1, ylim=None, xlim=None, show=True):
     '''
-    Plot connectivity estimation results.
+    Plot connectivity estimation results. Allows to plot your results
+    without using *Data* class.
     
     Args:
       *values* : numpy.array

connectivipy/mvar/fitting.py

@@ -9,7 +9,7 @@ def mvar_gen(Acf, n, omit=500):
     Generating data point from MVAR coefficiencs matrix *Acf*.
     Args:
       *Acf* : numpy.array
-          array of dimension kxkxp where *k* is number of channels and
+          array in shape of (p,k,k) where *k* is number of channels and
           *p* is a model order.
       *n* : int
           number of data points.
@@ -36,7 +36,7 @@ def mvar_gen_inst(Acf, n, omit=500):
     instantenous interaction not as in *mvar_gen* one data point lagged.
     Args:
       *Acf* : numpy.array
-          array of dimension kxkxp where *k* is number of channels and
+          array in shape of (p,k,k) where *k* is number of channels and
           *p* is a model order.
       *n* : int
           number of data points.

doc/conn.rst

@@ -3,9 +3,6 @@
 Connectvity
 ==================
 
-.. automodule:: __init__
-   :members:
-
 Connectivity methods classes.
 
 .. automodule:: conn

doc/data.rst

@@ -15,3 +15,8 @@ Data loading
 
 .. automodule:: load.loaders
    :members:
+
+Additional functions
+########
+
+.. autofunction:: __init__.plot_conn

examples/example2.py

@@ -0,0 +1,45 @@
+import numpy as np
+import matplotlib.pyplot as plt
+import connectivipy as cp
+from connectivipy.mvar.fitting import mvar_gen
+
+fs = 256.
+acf = np.zeros((3,3,3))
+# matrix shape meaning (p,k,k) k - number of channels,
+# p - order of mvar parameters
+
+acf[0,0,0] = 0.3
+acf[0,1,0] = 0.6
+acf[1,0,0] = 0.1
+acf[1,1,1] = 0.2
+acf[1,2,0] = 0.6
+acf[2,2,2] = 0.2
+acf[2,1,0] = 0.4
+
+# generate 3-channel signal from matrix above
+y = mvar_gen(acf,int(10e4))
+
+# assign static class cp.Mvar to variable mv
+mv = cp.Mvar
+
+# find best model order
+best, crit = mv._order_akaike(y,15,'vm')
+plt.plot(1+np.arange(len(crit)),crit,'g')
+plt.show()
+print best
+# here we know that this is 3 but in real-life cases
+# we are always uncertain about it
+
+# now let's fit parameters to the signal
+av, vf = mv.fit(y, best, 'vm')
+
+# and check whether values are correct +/- 0.01
+print np.allclose(acf, av, 0.01, 0.01)
+
+# now we can calculate Directed Transfer Function from the data
+dtf = cp.conn.DTF()
+dtfval = dtf.calculate(av, vf, 128)
+# all possible methods are visible in that dictionary:
+print cp.conn.conn_estim_dc.keys()
+
+cp.plot_conn(dtfval,'DTF values', fs)