Scatter Plot with Histograms using Python View for KNIME 4.7

[HeatherPikairos]

Hello KNIMErs

I’m back with another Matplotlib visualisation using the latest -Python View- node, so I’m sharing my workflow with you all

This time the workflow uses Python to create a Scatter Plot with Histograms:

I have used -Widget- nodes that allow the user to change the Title, Legend Title, Transparency etc.

You can find the workflow on the Hub here:

20230510 Pikairos Scatter Plot with Histograms using Python View for KNIME 4.7...

Best Wishes
Heather

[HeatherPikairos]

Sharing the Python Code for improved searching on Google etc

import knime.scripting.io as knio

from io import BytesIO
from matplotlib import colors
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns

# Only use numeric columns
data = knio.input_tables[0].to_pandas()

# Scatter plot with histograms function
def scatter_hist(x, y, ax, ax_histx, ax_histy):
    # no labels
    ax_histx.tick_params(axis="x", labelbottom=False)
    ax_histy.tick_params(axis="y", labelleft=False)

    # Useful data for setting colors
    c = data['Color']
    colors_set = np.unique(c)
    matrix = data.values

    # binwidth:
    binwidth = 0.25
    xymax = max(np.max(np.abs(x)), np.max(np.abs(y)))
    lim = (int(xymax/binwidth) + 1) * binwidth

    # For every color in the set, extract those that match 
    # the condition in a submatrix
    for current_color in colors_set:
        # Color choice
        condition = matrix[ :, 0] == current_color
        color_tab = "tab:" + current_color
        
        # Submatrix creation with only two columns of numbers
        color_submatrix = matrix[ np.nonzero( condition), 1:3]
        # Squeeze to remove a dimension
        color_submatrix = color_submatrix.squeeze()

        # Coordinates of the submatrix concerned
        x_mat = color_submatrix[ :, 0]
        y_mat = color_submatrix[ :, 1]

        # Scatter plot construction
        ax.scatter(x_mat, y_mat, c=color_tab, label=current_color, alpha = knio.flow_variables['transparency'])
        # Histogram construction
        bins = np.arange(-lim, lim + binwidth, binwidth)
        ax_histx.hist(x_mat.tolist(), bins = bins, density = True, alpha = knio.flow_variables['transparency'], color = current_color)
        ax_histy.hist(y_mat.tolist(), bins = bins, density = True, alpha = knio.flow_variables['transparency'], color = current_color, orientation='horizontal')
        # Density plot construction
        sns.kdeplot(x_mat.tolist(), ax=ax_histx, color = color_tab)
        sns.kdeplot(y_mat.tolist(), ax=ax_histy, color = color_tab, vertical = True)

# Plot size
left, width = 0.1, 0.65
bottom, height = 0.1, 0.65
spacing = 0.025

# Scatter plot size
rect_scatter = [left, bottom, width, height]
# Histogram sizes
rect_histx = [left, bottom + height + spacing, width, 0.2]
rect_histy = [left + width + spacing, bottom, 0.2, height]

# Start with a square Figure
fig = plt.figure(figsize=(8, 8))

# Adding plots to the principal plot
ax = fig.add_axes(rect_scatter)
ax_histx = fig.add_axes(rect_histx, sharex=ax)
ax_histy = fig.add_axes(rect_histy, sharey=ax)

# x-axis column
x = data['Dim1']
# y-axis column
y = data['Dim2']

# Use the previously defined function
scatter_hist(x, y, ax, ax_histx, ax_histy)

# Title
plt.title(knio.flow_variables['title'],loc = "center")

# Legend option
if knio.flow_variables['legend_required']:
    legend = ax.legend(loc = knio.flow_variables['legend_location'], 
    title=knio.flow_variables['legend_title'])
    ax.add_artist(legend)

# Grid option
if knio.flow_variables['grid_required']:
    ax.grid(True)

# Replace row ID by number
#data.index = range(0, len(data))
# Create buffer to write into
buffer = BytesIO()
# Create plot and write it into the buffer
fig.savefig(buffer, format='svg')
# The output is the content of the buffer
output_image = buffer.getvalue()

# Assign the figure to the output_view variable
knio.output_view = knio.view(fig)  # alternative: knio.view_matplotlib()