Review of functions in nilearn_plotting.py

[wingedRuslan]

While looking through the code in nilearn_plotting.py I've noted which parts of the code do and which parts of the code do not run. Opened this issue to share my findings.

1. graph_to_nilearn_array runs but would be better to have tests for this function.

2. plot_connectome_with_nilearn does not run.

• Besides having tiny typos (while calling graph_to_nilearn_array the parameters passed have different names), the execution of the function leads to the following error.

Error - Only length-1 arrays can be converted to Python scalars

TypeError                                 Traceback (most recent call last)
<ipython-input-17-3468803bff2b> in <module>()
----> 1 plot_connectome_with_nilearn(G)

~/anaconda3/lib/python3.6/site-packages/scona/nilearn_plotting.py in plot_connectome_with_nilearn(G, node_colour_att, node_colour, node_size_att, node_size, edge_attribute, edge_cmap, edge_vmin, edge_vmax, output_file, display_mode, figure, axes, title, annotate, black_bg, alpha, edge_kwargs, node_kwargs, colorbar)
     88         display_mode=display_mode, figure=figure, axes=axes, title=title,
     89         annotate=annotate, black_bg=black_bg, alpha=alpha,
---> 90         edge_kwargs=edge_kwargs, node_kwargs=node_kwargs, colorbar=colorbar)
     91 
     92 

~/anaconda3/lib/python3.6/site-packages/nilearn/plotting/img_plotting.py in plot_connectome(adjacency_matrix, node_coords, node_color, node_size, edge_cmap, edge_vmin, edge_vmax, edge_threshold, output_file, display_mode, figure, axes, title, annotate, black_bg, alpha, edge_kwargs, node_kwargs, colorbar)
   1263                       edge_threshold=edge_threshold,
   1264                       edge_kwargs=edge_kwargs, node_kwargs=node_kwargs,
-> 1265                       colorbar=colorbar)
   1266 
   1267     if output_file is not None:

~/anaconda3/lib/python3.6/site-packages/nilearn/plotting/displays.py in add_graph(self, adjacency_matrix, node_coords, node_color, node_size, edge_cmap, edge_vmin, edge_vmax, edge_threshold, edge_kwargs, node_kwargs, colorbar)
   1389                 ax._add_lines(line_coords, adjacency_matrix_values, edge_cmap,
   1390                               vmin=edge_vmin, vmax=edge_vmax,
-> 1391                               **edge_kwargs)
   1392             # To obtain the brain left view, we simply invert the x axis
   1393             if ax.direction == 'l':

~/anaconda3/lib/python3.6/site-packages/nilearn/plotting/displays.py in _add_lines(self, line_coords, line_values, cmap, vmin, vmax, **kwargs)
    420             this_kwargs.update(kwargs)
    421             xdata, ydata = start_end_point_2d.T
--> 422             line = lines.Line2D(xdata, ydata, **this_kwargs)
    423             self.ax.add_line(line)
    424 

~/anaconda3/lib/python3.6/site-packages/matplotlib/lines.py in __init__(self, xdata, ydata, linewidth, linestyle, color, marker, markersize, markeredgewidth, markeredgecolor, markerfacecolor, markerfacecoloralt, fillstyle, antialiased, dash_capstyle, solid_capstyle, dash_joinstyle, solid_joinstyle, pickradius, drawstyle, markevery, **kwargs)
    396         self.set_linestyle(linestyle)
    397         self.set_drawstyle(drawstyle)
--> 398         self.set_linewidth(linewidth)
    399 
    400         self._color = None

~/anaconda3/lib/python3.6/site-packages/matplotlib/lines.py in set_linewidth(self, w)
   1004         ACCEPTS: float value in points
   1005         """
-> 1006         w = float(w)
   1007 
   1008         if self._linewidth != w:

~/anaconda3/lib/python3.6/site-packages/numpy/ma/core.py in __float__(self)
   4294         """
   4295         if self.size > 1:
-> 4296             raise TypeError("Only length-1 arrays can be converted "
   4297                             "to Python scalars")
   4298         elif self._mask:

TypeError: Only length-1 arrays can be converted to Python scalars

3. view_connectome_with_nilearn works fine. But while running the following error appears:

Error - iopub_data_rate_limit

IOPub data rate exceeded.
The notebook server will temporarily stop sending output
to the client in order to avoid crashing it.
To change this limit, set the config variable
`--NotebookApp.iopub_data_rate_limit`.

This is a known issue with jupyter notebook - iopub rate limits are too low by default, for visualization-heavy projects. To solve this a user should change jupyter's config variable iopub_data_rate_limit. One possible solution is to run the notebook with the following command (changing the parameter to arbitrary big number):
jupyter notebook --NotebookApp.iopub_data_rate_limit=10000000
After running a notebook as stated above, the function is successfully executed and produces a nice figure.

4. view_markers_with_nilearn runs and everything is fine.

---

In order to run the above-mentioned functions, I have used this code from available tutorials to make the required input argument - BrainNetwork Graph:

import numpy as np
import networkx as nx
import scona as scn
import scona.datasets as datasets

# Read in sample data from the NSPN WhitakerVertes PNAS 2016 paper.
df, names, covars, centroids = datasets.NSPN_WhitakerVertes_PNAS2016.import_data()

# calculate residuals of the matrix df for the columns of names
df_res = scn.create_residuals_df(df, names)

# create a correlation matrix over the columns of df_res
M = scn.create_corrmat(df_res, method='pearson')

# Initialise a weighted graph G from the correlation matrix M
G = scn.BrainNetwork(network=M, parcellation=names, centroids=centroids)

For successfully executed functions I have created a jupyter notebook to demonstrate the outputs.