ModelOriented / ModelOriented/DALEX

`plot.model_diagnostics()` plots tiny points with `base_line_size = 0` in CRAN version 2.5.3

Open
#581 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1.5k
Forks
172
PR merge metrics
No merged PRs in 30d

Description

The current version 2.5.3 on CRAN generates some strange plot.model_diagnostics() charts. The issue seems to be theme_bw(base_line_size = 0) in theme_drwhy(). I'm not sure why this is different between CRAN and GH though. Reprex:

library(DALEX)

# Example from Explanatory Model Analysis, 19 Residual-diagnostics Plots, 19.4 Example: apartment-prices data
# https://ema.drwhy.ai

# Make randomForest model
library("randomForest")
set.seed(72)

model_apart_rf <- randomForest(m2.price ~ ., data = apartments)

explain_apart_rf <- DALEX::explain(
  model = model_apart_rf,
  verbose = F,
  data = apartments_test[, -1],
  y = apartments_test$m2.price,
  label = "Random Forest"
)

diag_rf <- model_diagnostics(explain_apart_rf)

plot(diag_rf)

CRAN v2.5.3 results

Image

GitHub v2.5.3 results

Image

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by inspecting the theme_drwhy() and plot.model_diagnostics() implementations, then reproduce the provided randomForest example with CRAN 2.5.3 and the GitHub version. Trace the base_line_size setting and confirm that the resulting diagnostic plots no longer render tiny points or otherwise differ unexpectedly.

Written by the indexing model from the issue text.

Assessment

Tech stack
r
Domain
data-visualization, machine-learning
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Clearly specified
Newbie friendliness
52/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.