how to draw each node by networkx?

I know there's a function, nx.draw_networkx_nodes(G, pos, node_size=200, node_color='#00b4d9') can draw all nodes in the network. However, each of my node have a different color,dataset and code like that,

df_name = ['A','B','C','D','E'] val_idx = [1,2,3,4,5] pred_pca_spec = [1,2,0,1,4] #Note! below is rgb color value of each sample val_rgb = [[130, 89, 34],[170, 133, 75],[126, 85, 17],[120, 81, 37],[162, 126, 65]] G = nx.Graph() G.add_node('no.1') pos = nx.spring_layout(G) for i in range(0, len(df_name)): if pred_pca_spec[i] == 1: G.add_edge('no.1', int(val_idx[i])) # I want to set color in this line at same time,like ######G.add_edge('no.1', node_value = int(val_idx[i]), node_color = '#%02x%02x%02x' % tuple(val_rgb[i]))##### nx.draw_networkx(G) plt.show() 

Is there some function can achieve it? Note!Because dataset are dyes textile samples so I must show true color of them, only can use the rgb color space, not

colors = ["r", "g", "b", "y", "w"]

please!

1 Answer

I'm still not sure, what exactly your problem is or where you struggle or why you want to add the color within the for loop. However, I hope the following code solves your problem. Using Color using RGB value in Matplotlib I casted your RGB to values between 0 and 1.

import networkx as nx import matplotlib.pylab as plt df_name = ["A", "B", "C", "D", "E"] val_idx = [1, 2, 3, 4, 5] pred_pca_spec = [1, 2, 0, 1, 4] val_rgb = [[130, 89, 34], [170, 133, 75], [126, 85, 17], [120, 81, 37], [162, 126, 65], ] val_rgb = [(r / 255, g / 255, b / 255) for r, g, b in val_rgb] G = nx.Graph() G.add_node('no.1') G.nodes['no.1']["color"] = tuple(val_rgb[0]) # or whatever color you like pos = nx.spring_layout(G) for i in range(0, len(df_name)): if pred_pca_spec[i] == 1: node = int(val_idx[i]) G.add_edge('no.1', node) G.nodes[node]["color"] = val_rgb[i] nx.draw_networkx(G, node_color=[G.nodes[node]["color"] for node in G]) plt.show() 

Result