How to plot large dataset of Shapely LineString?

📊 Plotly Python
1

I’ve been given a DXF file at work to analyse it, and they want me to plot it as well. This file contains a layout of a private area. The data looks like this:

Layer   PaperSpace    SubClasses              Linetype  EntityHandle Text  geometry
860     0       None  AcDbEntity:AcDbPolyline     None          59E  None  LINESTRING (441.981 186.290, 441.981 189.994, ...
861     0       None  AcDbEntity:AcDbPolyline     None          59F  None  LINESTRING (441.981 193.740, 441.981 190.036, ...
862     0       None  AcDbEntity:AcDbPolyline     None          5A0  None      LINESTRING (441.790 189.994, 441.790 190.036)
863     0       None  AcDbEntity:AcDbPolyline     None          5A1  None      LINESTRING (441.790 190.015, 441.790 190.015)
864     0       None  AcDbEntity:AcDbPolyline     None          5A2  None      LINESTRING (441.790 186.544, 441.790 186.544)

The whole dataset has 397.535 records. The column I’m interested is the geometry one, which is full of Shapely LineString objects. Even if I can plot it without issues with GeoPandas plot, I want to be able to plot the geometry column with Plotly (since this allow me to zoom in, out, see values when hovering over the plot, etc.). However, I cannot plot the whole dataset without having VSCode or the browser (I tried Jupyter Notebook on web and also tried to open an html file generated with fig.write_html, with no success) to crash.

This is the whole code:

import geopandas as gpd
import pandas as pd
import plotly.graph_objects as go
    
layout_dfs = []
test = gpd.read_file('/test_layout_file.dxf')
    
polyline_df = test[test.SubClasses == 'AcDbEntity:AcDbPolyline']
# polyline_df.head()
    
plot_data = []
for polygon in polyline_df.geometry[122:10000]:
    x, y = polygon.xy
    #print(y)
    plot_data.append(
        go.Scatter(
            x=list(x[1:]),
            y=list(y[1:]),
            fill="toself",  # Fills the polygon
            mode="none",  # Hides markers
            showlegend=False
         )
    )
    
layout = go.Layout(
    autosize=False,
    width=1300,
    height=800,
    margin=go.layout.Margin(l=50, r=50, b=100, t=100, pad=4),
)
    
# Create the figure and display
fig = go.Figure(data=plot_data, layout=layout)
fig.show()
2

Hey @jmg241 welcome to the forums!

Just to make this clear: Are you able to render a fraction of the data without problems using plotly, right?

3

Yes, exactly. Once I try to plot anything beyond approximately 100.000 records, the plot doesn’t render.

4

Have you tried the.explore from geopandas? Might be worth checking whether this displays it, it also gives you an interactive map, follium-based if memory serves.

5

I’ll have a look at it, thanks