Coordinate Systems¶

Coordinate systems control how x and y positions are mapped to the plot.

In [1]:

import pandas as pd
import numpy as np
from ggplotly import *

# Sample data
np.random.seed(42)
df = pd.DataFrame({
    'x': np.random.rand(100) * 100,
    'y': np.random.rand(100) * 100,
    'category': np.random.choice(['A', 'B', 'C', 'D'], 100)
})

# Bar chart data
bar_df = pd.DataFrame({
    'category': ['A', 'B', 'C', 'D'],
    'value': [25, 40, 35, 30]
})

# Distribution data
dist_df = pd.DataFrame({
    'category': np.repeat(['A', 'B', 'C'], 50),
    'value': np.concatenate([
        np.random.normal(0, 1, 50),
        np.random.normal(2, 1.5, 50),
        np.random.normal(-1, 0.8, 50)
    ])
})

import pandas as pd import numpy as np from ggplotly import * # Sample data np.random.seed(42) df = pd.DataFrame({ 'x': np.random.rand(100) * 100, 'y': np.random.rand(100) * 100, 'category': np.random.choice(['A', 'B', 'C', 'D'], 100) }) # Bar chart data bar_df = pd.DataFrame({ 'category': ['A', 'B', 'C', 'D'], 'value': [25, 40, 35, 30] }) # Distribution data dist_df = pd.DataFrame({ 'category': np.repeat(['A', 'B', 'C'], 50), 'value': np.concatenate([ np.random.normal(0, 1, 50), np.random.normal(2, 1.5, 50), np.random.normal(-1, 0.8, 50) ]) })

coord_cartesian¶

The default Cartesian coordinate system. Use it to zoom without clipping data.

In [2]:

# Zoom to a region (data outside is still used for calculations)
ggplot(df, aes(x='x', y='y')) + geom_point() + coord_cartesian(xlim=(20, 80), ylim=(20, 80))

# Zoom to a region (data outside is still used for calculations) ggplot(df, aes(x='x', y='y')) + geom_point() + coord_cartesian(xlim=(20, 80), ylim=(20, 80))

Out[2]:

xlim/ylim vs coord_cartesian: xlim() and ylim() filter the data before plotting. coord_cartesian() zooms the view without removing data points. This matters for statistical calculations like geom_smooth.

coord_fixed¶

Fixed aspect ratio coordinates. Essential for maps and any plot where x and y must have equal scaling.

In [3]:

# Square data should appear as a square with ratio=1
square_df = pd.DataFrame({'x': [0, 1, 1, 0, 0], 'y': [0, 0, 1, 1, 0]})
ggplot(square_df, aes(x='x', y='y')) + geom_path(size=2) + geom_point(size=10) + coord_fixed()

# Square data should appear as a square with ratio=1 square_df = pd.DataFrame({'x': [0, 1, 1, 0, 0], 'y': [0, 0, 1, 1, 0]}) ggplot(square_df, aes(x='x', y='y')) + geom_path(size=2) + geom_point(size=10) + coord_fixed()

Out[3]:

In [4]:

# With ratio=2, the square appears as a tall rectangle (y is stretched)
ggplot(square_df, aes(x='x', y='y')) + geom_path(size=2) + geom_point(size=10) + coord_fixed(ratio=2)

# With ratio=2, the square appears as a tall rectangle (y is stretched) ggplot(square_df, aes(x='x', y='y')) + geom_path(size=2) + geom_point(size=10) + coord_fixed(ratio=2)

Out[4]:

coord_flip¶

Flip x and y axes. Useful for horizontal bar charts and boxplots.

In [5]:

# Horizontal bar chart
ggplot(bar_df, aes(x='category', y='value')) + geom_bar(stat='identity') + coord_flip()

# Horizontal bar chart ggplot(bar_df, aes(x='category', y='value')) + geom_bar(stat='identity') + coord_flip()

Out[5]:

In [6]:

# Horizontal boxplot
ggplot(dist_df, aes(x='category', y='value')) + geom_boxplot() + coord_flip()

# Horizontal boxplot ggplot(dist_df, aes(x='category', y='value')) + geom_boxplot() + coord_flip()

Out[6]:

coord_polar¶

Polar coordinates for pie charts and radial plots.

coord_polar Parameters¶

coord_polar(
    theta='x',      # Variable mapped to angle ('x' or 'y')
    start=0,        # Starting angle in radians
    direction=1     # 1 = clockwise, -1 = counter-clockwise
)

coord_sf¶

For geographic/spatial data with proper map projections. Requires geopandas.

import geopandas as gpd

# Load geographic data
world = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))

# Plot with projection
ggplot(world) + geom_sf() + coord_sf(crs='EPSG:4326')

Available Projections¶

• 'mercator' - Web Mercator (Google Maps style)
• 'albers usa' - Albers USA (good for US maps)
• 'orthographic' - Globe view
• 'natural earth' - Natural Earth projection
• 'robinson' - Robinson projection

Coordinate Reference¶

Function	Description
coord_cartesian	Default Cartesian, zoom without clipping
coord_fixed	Fixed aspect ratio (ratio=1 for equal scaling)
coord_flip	Flip x and y axes
coord_polar	Polar coordinates
coord_sf	Geographic projections

coord_polar Parameters¶

coord_polar(
    theta='x',      # Variable mapped to angle ('x' or 'y')
    start=0,        # Starting angle in radians
    direction=1     # 1 = clockwise, -1 = counter-clockwise
)

coord_sf¶

For geographic/spatial data with proper map projections. Requires geopandas.

import geopandas as gpd

# Load geographic data
world = gpd.read_file(gpd.datasets.get_path('naturalearth_lowres'))

# Plot with projection
ggplot(world) + geom_sf() + coord_sf(crs='EPSG:4326')

Available Projections¶

• 'mercator' - Web Mercator (Google Maps style)
• 'albers usa' - Albers USA (good for US maps)
• 'orthographic' - Globe view
• 'natural earth' - Natural Earth projection
• 'robinson' - Robinson projection

Coordinate Reference¶

Function	Description
coord_cartesian	Default Cartesian, zoom without clipping
coord_flip	Flip x and y axes
coord_polar	Polar coordinates
coord_sf	Geographic projections