plotly 4.13.0

v4.13.0 - Magical Error Messages and Documentation, plus Speedups and Faceted Maps

on Python PyPI

See the fully release announcement at https://community.plotly.com/t/announcing-plotly-py-4-13-magical-error-messages-and-documentation-plus-speedups-and-faceted-maps/47497

Added

• px.choropleth, px.scatter_geo and px.line_geo now support faceting as well as fitbounds and basemap_visible 2923
• px.scatter_geo and px.line_geo now support geojson/featureidkey input 2923
• px.scatter_geo now supports symbol 2923
• go.Figure now has a set_subplots method to set subplots on an already
  existing figure. 2866
• Added Turbo colorscale and fancier swatch display functions
  2882
• A utility function image_array_to_data_uri has been added in
  plotly.utils, in order to transform NumPy arrays to data b64 URIs (which
  can be passed to the source parameter of go.Image, or to layout images).
  2879
• the selector argument to updater/selector functions now accepts ints and strs
  2894

Updated

• the JSON serialization of plotly figures has been accelerated thanks to a
  different handling of Infinity and NaN values. For example, a figure with a
  1000x1000 Heatmap should now serialize 2x faster. 2880
• Coding mistakes with "magic underscores" now return significantly more ergonomic error
  messages 2843
• Error messages related to impossible subplot geometries are now much more helpful
  2897

Fixed

• px.scatter_geo support for text is fixed 2923
• the x and y parameters of px.imshow are now used also in the case where
  an Image trace is used (for RGB data or with binary_string=True). However,
  only numerical values are accepted (while the Heatmap trace allows date or
  string values for x and y). 2761