Updating layout, axes and legends

Almost every chart needs tweaks after creation, and these update methods are the standard way to make them.

Plotly figures are edited after creation through update_* methods that take keyword arguments and return the figure, so they chain:

fig.update_layout(title_text="Revenue", height=400, legend_title_text="Region")
fig.update_xaxes(title_text="Month", tickangle=-45)
fig.update_yaxes(tickprefix="$", tickformat=",.0f", rangemode="tozero")

Magic underscores let you reach nested attributes in one keyword: title_font_size=20 is the same as title=dict(font=dict(size=20)). Updates are merged, not replaced, so changing title_font_size keeps the title text.

update_xaxes and update_yaxes apply to all axes at once, which matters for faceted figures with many axes; pass row=, col= or selector= to target some of them.

Frequently used settings:

axes: range=[lo, hi], type="log", tickformat (d3-format, e.g. ".1%", ",.0f"), dtick, showgrid, zeroline, matches=None to unlink facet axes;
legend: orientation="h", x, y, xanchor, yanchor, showlegend=False;
margin=dict(l=40, r=20, t=50, b=40), width, height;
hovermode="x unified" for one tooltip listing all traces at an x.

Assigning attributes directly also works: fig.layout.title.text = "Revenue".

Example

import plotly.express as px

df = px.data.stocks()
fig = px.line(df, x="date", y=["GOOG", "MSFT"])

(fig.update_layout(title_text="Stocks normalized to 1 at start",
                   title_font_size=20, legend_title_text="Ticker",
                   legend=dict(orientation="h", y=1.02, x=0, yanchor="bottom"),
                   hovermode="x unified", height=420)
    .update_xaxes(title_text=None, dtick="M3", tickformat="%b %Y")
    .update_yaxes(tickformat=".0%", title_text="Relative price"))

print(fig.layout.title.text)          # Stocks normalized to 1 at start
print(fig.layout.title.font.size)     # 20
print(fig.layout.yaxis.tickformat)    # .0%
fig.write_html("stocks.html")

Remember

update_layout, update_xaxes, update_yaxes merge changes and chain
Magic underscores reach nested attributes: title_font_size=20
Tick formats use d3-format strings such as ,.0f and .1%

Watch out for

Calling fig.update_layout(xaxis_range=...) on a faceted figure and only changing the first panel; use update_xaxes.

Check yourself

What is `fig.update_layout(title_font_color='gray')` equivalent to?

Official docs: plotly.com/python/creating-and-updating-figures/

Example

Remember

Watch out for

Check yourself

Builds on

Unlocks