Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix issues with explicitly transformed recipes in DataInspector #3553

Draft
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

ffreyer
Copy link
Collaborator

@ffreyer ffreyer commented Jan 14, 2024

Description

Fixes #3526

With a plot structure like

recipe_plot
  primitive

where recipe_plot does not implement it's own show_data method, DataInspector will fall back on show_data for primitive. There it will assume that the input data of the primitive plot is in data space. If the recipe_plot handles the transform function directly, passing the result to primitive, the assumption that the input data is in data space is not correct. So you get incorrect labels.

This should affect every recipe plot that applies transform_func before creating child plots. That is:

  • h/vlines
  • h/vspan
  • maybe barplot
  • bracket not inspectable at all?
  • contours (also missplaced)
  • error/rangebars (also missplaced)
  • poly
  • text with lines (tex) not inspectable at all?
  • tooltip (skipping this since it's unlikely to get inspected)
  • triplot (also bad limits)
  • voronoiplot (also bad limits)

Type of change

  • Bug fix (non-breaking change which fixes an issue)

Checklist

  • Added an entry in NEWS.md (for new features and breaking changes)
  • Added or changed relevant sections in the documentation
  • Added unit tests for new algorithms, conversion methods, etc.
  • Added reference image tests for new plotting functions, recipes, visual options, etc.

@MakieBot
Copy link
Collaborator

MakieBot commented Jan 14, 2024

Compile Times benchmark

Note, that these numbers may fluctuate on the CI servers, so take them with a grain of salt. All benchmark results are based on the mean time and negative percent mean faster than the base branch. Note, that GLMakie + WGLMakie run on an emulated GPU, so the runtime benchmark is much slower. Results are from running:

using_time = @ctime using Backend
# Compile time
create_time = @ctime fig = scatter(1:4; color=1:4, colormap=:turbo, markersize=20, visible=true)
display_time = @ctime Makie.colorbuffer(display(fig))
# Runtime
create_time = @benchmark fig = scatter(1:4; color=1:4, colormap=:turbo, markersize=20, visible=true)
display_time = @benchmark Makie.colorbuffer(fig)
using create display create display
GLMakie 3.37s (3.33, 3.38) 0.02+- 386.40ms (371.86, 397.63) 9.46+- 469.57ms (462.32, 501.40) 14.09+- 7.00ms (6.92, 7.08) 0.06+- 25.16ms (25.09, 25.28) 0.06+-
master 3.36s (3.35, 3.37) 0.01+- 394.16ms (389.44, 399.07) 3.29+- 462.31ms (459.51, 465.63) 2.25+- 7.03ms (6.96, 7.10) 0.05+- 25.19ms (25.10, 25.27) 0.07+-
evaluation 1.00x invariant, 0.0s (0.25d, 0.65p, 0.01std) 1.02x invariant, -7.76ms (-1.10d, 0.08p, 6.37std) 0.98x invariant, 7.26ms (0.72d, 0.22p, 8.17std) 1.01x invariant, -0.04ms (-0.65d, 0.25p, 0.05std) 1.00x invariant, -0.03ms (-0.45d, 0.42p, 0.06std)
CairoMakie 3.06s (3.03, 3.09) 0.03+- 321.41ms (316.11, 331.78) 6.03+- 140.37ms (138.78, 143.90) 1.72+- 7.20ms (7.17, 7.25) 0.03+- 609.54μs (604.51, 614.89) 3.71+-
master 3.04s (3.02, 3.09) 0.02+- 320.18ms (316.84, 326.16) 3.08+- 140.13ms (138.98, 140.99) 0.77+- 7.20ms (7.14, 7.28) 0.05+- 605.22μs (600.89, 607.96) 2.65+-
evaluation 0.99x invariant, 0.02s (0.70d, 0.21p, 0.03std) 1.00x invariant, 1.23ms (0.26d, 0.64p, 4.55std) 1.00x invariant, 0.24ms (0.18d, 0.75p, 1.24std) 1.00x invariant, 0.0ms (0.05d, 0.93p, 0.04std) 0.99x slower X, 4.32μs (1.34d, 0.03p, 3.18std)
WGLMakie 3.64s (3.60, 3.68) 0.02+- 317.85ms (316.63, 319.20) 0.81+- 8.80s (8.66, 9.00) 0.12+- 9.00ms (8.90, 9.13) 0.10+- 73.54ms (66.79, 79.40) 4.92+-
master 3.62s (3.60, 3.64) 0.02+- 320.17ms (317.15, 323.16) 2.27+- 8.77s (8.72, 8.86) 0.05+- 9.26ms (8.87, 10.78) 0.68+- 72.68ms (67.29, 80.34) 4.79+-
evaluation 0.99x invariant, 0.02s (0.92d, 0.12p, 0.02std) 1.01x faster ✓, -2.33ms (-1.36d, 0.04p, 1.54std) 1.00x invariant, 0.03s (0.34d, 0.54p, 0.08std) 1.03x invariant, -0.26ms (-0.53d, 0.36p, 0.39std) 0.99x invariant, 0.86ms (0.18d, 0.74p, 4.86std)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

DataInspector still shows transformed coordinated for log scale, hline
2 participants