Changes in Version 7 in Python

Guide to changes in version 7 of Plotly.py and how to migrate from version 6

Plotly Studio: Transform any dataset into an interactive data application in minutes with AI. Try Plotly Studio now.

This page outlines the changes in Plotly.py version 7 and cases where you may need to update your charts or the tools you use for working with Plotly.py.

Version 7 upgrades the bundled Plotly.js from version 3 to version 4, so it includes the Plotly.js changes as well as the Plotly.py ones. Most figures render without any changes. The sections below cover the cases where you need to update your code.

Removed Traces¶

The Mapbox-based traces were deprecated in version 6 and have been removed in version 7. Use the MapLibre-based traces instead, which have the same attributes.

Removed	Use instead
`go.Scattermapbox`	`go.Scattermap`
`go.Choroplethmapbox`	`go.Choroplethmap`
`go.Densitymapbox`	`go.Densitymap`
`px.scatter_mapbox`	`px.scatter_map`
`px.line_mapbox`	`px.line_map`
`px.choropleth_mapbox`	`px.choropleth_map`
`px.density_mapbox`	`px.density_map`

In [1]:

import plotly.graph_objects as go

fig = go.Figure(go.Scattermap(
    lon=[-73.57, -79.38, -123.12],
    lat=[45.50, 43.65, 49.28],
    mode="markers",
    marker=dict(size=12),
))

fig.show()

See Migrate to MapLibre for the full list of changes when moving from the Mapbox traces.

Removed Attributes¶

`layout.mapbox`¶

The mapbox subplot has been removed. Use layout.map, which takes the same attributes. In Plotly Express, mapbox_style becomes map_style.

`stream`¶

The stream attribute has been removed from all traces. It configured the Chart Studio streaming service, which is no longer available.

`*src` attributes¶

Attributes ending in src (xsrc, ysrc, textsrc, marker.colorsrc, and so on) have been removed, along with layout.hidesources. They referenced data sources in Chart Studio.

Chart Studio configuration options¶

The showLink, link_text, sendData, showSources, and showEditInChartStudio configuration options have been removed. To let users share a chart, use showSendToCloud, which is enabled by default in version 7 and adds a "Share chart…" button to the mode bar.

Removed Figure Factories¶

The deprecated figure factories create_2d_density, create_annotated_heatmap, create_bullet, create_candlestick, create_choropleth, create_distplot, create_facet_grid, create_gantt, create_hexbin_mapbox, create_ohlc, create_scatterplotmatrix, and create_violin have been removed. Figure Factories lists the recommended alternative for each one.

Static Image Export¶

Support for Kaleido versions earlier than 1.0.0 has been removed, along with support for Orca. The engine argument has been removed from fig.write_image(), fig.to_image(), pio.write_image(), pio.write_images(), pio.to_image(), pio.full_figure_for_development(), and from renderer constructors. See Static Image Export.

Changed Defaults¶

Maps fit to your data¶

geo subplots (used by go.Scattergeo and go.Choropleth) now fit their initial view to the locations they plot: layout.geo.fitbounds defaults to "locations" instead of False. Tile-based map subplots do the same, through a new layout.map.fitbounds attribute that also defaults to "locations".

Figures that already set a view render unchanged. To get the previous world view, set fitbounds=False:

In [2]:

import plotly.graph_objects as go

fig = go.Figure(go.Choropleth(locations=["CAN", "MEX"], z=[1, 2], showscale=False))

fig.update_layout(geo=dict(fitbounds=False))

fig.show()

Setting your own view attributes also turns auto-fitting off. On a geo subplot, center and projection.scale always do this, and projection.rotation, lonaxis.range, and lataxis.range do except on a scoped subplot with a clipped projection. On a map subplot, center or zoom does it.

An axis that overlays another (yaxis2.overlaying="y") now defaults tickmode to "sync": it draws ticks and gridlines at the base axis's positions, labeled from its own range, so the two axes share one grid. Set tickmode="auto" on the overlaying axis for the previous behavior of two independent grids.

Scatter plot matrix axes are linked¶

go.Splom now defaults axis.matches to True, so axes in the same row and column are linked and pan and zoom together. Plotly Express already set this on the figures it generated. Set matches=False on the relevant axes for the previous behavior.

Color Parsing¶

Plotly.js now parses colors according to the CSS Color 4 specification. Four color string formats that previously worked no longer produce the same color, and an unparseable color falls back to the attribute's default:

No longer works	Example	Use instead
The `hsv()` color function	`"hsv(200, 80%, 80%)"`	`"hsl(200, 67%, 47%)"`, `"hwb()"`, hexadecimal, or `"rgb()"`
Comma separated saturation and lightness values without percent units	`"hsl(0, 100, 40)"`	`"hsl(0, 100%, 40%)"`, or the space separated `"hsl(0 100 40)"`
Channel values given as fractions of 1	`"rgb(0.5, 0.5, 0.5)"` (now renders as near black)	`"rgb(128, 128, 128)"`
Hexadecimal values without a leading `#`	`"fff"`	`"#fff"`

These rules apply to color strings only. Numeric arrays used with a colorscale are unaffected. Automatically computed contrast colors, such as text on heatmap cells, can also shift slightly around mid-luminance backgrounds.

Separately, plotly.colors.hex_to_rgb now parses 3-digit shorthand hexadecimal colors such as #FFF correctly.

Country Names¶

locationmode="country names" is now resolved by country-iso-search rather than a set of regular expressions. It accepts more forms — ISO-3166 codes, UN M49 numeric codes, flag emoji, historical and native names such as "Burma" and "Türkiye" — and ignores case, accents, and punctuation. It rejects partial phrases that previously matched by accident. Unrecognized names are logged and their locations are skipped. Use locationmode="ISO-3" with ISO codes if you need certainty.

Tile Map Rendering¶

Three rendering details changed for go.Scattermap:

Marker icons are tinted with marker.color. Previously they were always black. Pass marker=dict(color="black") for the previous appearance.
The Maki icon set was updated from version 2.1 to 8.2. A few icon names were removed between those versions.
Legend swatches always draw a circle, regardless of marker.symbol.

Sankey Layout¶

The Sankey layout algorithm was updated (@plotly/d3-sankey 0.7.2 to 0.12.3). Node positions and link paths shift slightly for the same data, with links tending to cross less.

Two new Sankey attributes let you override the automatic ordering: node.sort and link.sort.

For both attributes, the default value "auto" reorders nodes within a column, and links within a node, to reduce crossings. "input" keeps the order given in node.label and in link.source / link.target, which is useful when the order carries meaning or you need a layout that is stable across renders.

In [3]:

import plotly.graph_objects as go

fig = go.Figure(go.Sankey(
    node=dict(label=["A", "B", "C", "D", "Z", "Y", "X", "W"], pad=15, thickness=20,
              sort="input"),
    link=dict(source=[0, 1, 2, 3], target=[4, 5, 6, 7], value=[4, 3, 2, 1],
              sort="input"),
))

fig.show()

A third attribute, direction, controls the flow along the orientation axis. The default value of "forward" results in a chart which flows from left to right horizontally, or top to bottom vertically. Set it to "reversed" to create a chart which flows from right to left or bottom to top. See Sankey Diagram.

New Traces and Attributes¶

`go.Quiver`¶

Version 7 adds a Quiver trace type for 2D vector fields, which supercedes the create_quiver figure factory as the recommended approach for creating quiver plots. See Quiver Plots for more detail.

In [4]:

import plotly.graph_objects as go

fig = go.Figure(go.Quiver(x=[0, 1, 2, 0, 1, 2], y=[0, 0, 0, 1, 1, 1],
                          u=[1, 0.5, 0, 0.5, 0, -0.5], v=[0, 0.5, 1, 0.5, 1, 0.5]))

fig.show()

Geo zoom limits¶

New attributes layout.geo.projection.minscale and maxscale clamp how far users can zoom a geo subplot. Both are multipliers of projection.scale:

In [5]:

import plotly.graph_objects as go

fig = go.Figure(go.Scattergeo(lon=[2.35, 13.4, 12.5, -3.7], lat=[48.86, 52.52, 41.9, 40.42],
                              mode="markers"))

fig.update_layout(geo=dict(scope="europe",
                           projection=dict(scale=1, minscale=0.5, maxscale=4)))

fig.show()

MathJax¶

Plotly.js now supports MathJax version 3 and version 4 for rendering LaTeX, and support for MathJax version 2 has been removed. If you render figures in an environment that loads MathJax itself, such as a notebook or a custom HTML page, make sure it loads version 3 or later.

What About Dash?¶

Dash is an open-source framework for building analytical applications, with no Javascript required, and it is tightly integrated with the Plotly graphing library.

Learn about how to install Dash at https://dash.plot.ly/installation.

Everywhere in this page that you see fig.show(), you can display the same figure in a Dash application by passing it to the figure argument of the Graph component from the built-in dash_core_components package like this:

import plotly.graph_objects as go # or plotly.express as px
fig = go.Figure() # or any Plotly Express function e.g. px.bar(...)
# fig.add_trace( ... )
# fig.update_layout( ... )

from dash import Dash, dcc, html

app = Dash()
app.layout = html.Div([
    dcc.Graph(figure=fig)
])

app.run(debug=True, use_reloader=False)  # Turn off reloader if inside Jupyter