CRAN Package Check Results for Package DALEX

Last updated on 2020-07-06 21:50:13 CEST.

Flavor Version Tinstall Tcheck Ttotal Status Flags
r-devel-linux-x86_64-debian-clang 1.3.0 6.31 88.71 95.02 OK
r-devel-linux-x86_64-debian-gcc 1.3.0 5.83 68.88 74.71 OK
r-devel-linux-x86_64-fedora-clang 1.3.0 117.43 OK
r-devel-linux-x86_64-fedora-gcc 1.3.0 110.98 ERROR
r-devel-windows-ix86+x86_64 1.3.0 17.00 102.00 119.00 OK
r-patched-linux-x86_64 1.3.0 6.51 87.45 93.96 OK
r-patched-solaris-x86 1.3.0 261.90 OK
r-release-linux-x86_64 1.3.0 6.23 86.05 92.28 OK
r-release-osx-x86_64 1.2.1 OK
r-release-windows-ix86+x86_64 1.3.0 27.00 138.00 165.00 OK
r-oldrel-osx-x86_64 1.3.0 OK
r-oldrel-windows-ix86+x86_64 1.2.1 27.00 126.00 153.00 OK

Check Details

Version: 1.3.0
Check: tests
Result: ERROR
     Running ‘testthat.R’ [35s/49s]
    Running the tests in ‘tests/testthat.R’ failed.
    Complete output:
     > library(testthat)
     > library(DALEX)
     Welcome to DALEX (version: 1.3.0).
     Find examples and detailed introduction at: https://pbiecek.github.io/ema/
    
     >
     > test_check("DALEX")
     Package: ranger
     Package version: 0.12.1
     Task type: regressionModel label: ranger
     Model class: ranger
     Data head :
     construction.year surface floor no.rooms district
     1 1953 25 3 1 Srodmiescie
     2 1992 143 9 5 Bielany
     Preparation of a new explainer is initiated
     -> model label : ranger ( <1b>[33m default <1b>[39m )
     -> data : 7847 rows 6 cols
     -> target variable : 7847 values
     -> target variable : Please note that 'y' is a factor. ( <1b>[31m WARNING <1b>[39m )
     -> target variable : Consider changing the 'y' to a logical or numerical vector.
     -> target variable : Otherwise I will not be able to calculate residuals or loss function.
     -> predict function : yhat.ranger will be used ( <1b>[33m default <1b>[39m )
     -> predicted values : predict function returns multiple columns: 3 ( <1b>[31m WARNING <1b>[39m ) some of functionalities may not work
     -> model_info : package ranger , ver. 0.12.1 , task multiclass ( <1b>[33m default <1b>[39m )
     -> residual function : difference between 1 and probability of true class ( <1b>[33m default <1b>[39m )
     -> residuals : numerical, min = 0 , mean = 0.2774409 , max = 0.883016
     <1b>[32m A new explainer has been created! <1b>[39m
     Measures for: regression
     mse : 79358.95
     rmse : 281.7072
     r2 : 0.900389
     mad : 211.999
    
     Residuals:
     0% 10% 20% 30% 40% 50% 60% 70%
     -249.1437 -223.8050 -208.7184 -195.0442 -184.4288 -175.5123 -161.8545 366.1237
     80% 90% 100%
     396.9122 424.0089 470.8357
     The number of important variables for ranger's prediction is 4 out of 6.
     Variables district, surface, floor have the highest importantance.
     ── 1. Failure: parsnip (@test_yhat_model_info.R#164) ──────────────────────────
     explainer_classif_parsnip$y_hat inherits from `NULL` not `numeric`.
    
     ── 2. Failure: parsnip (@test_yhat_model_info.R#166) ──────────────────────────
     explainer_regr_parsnip$y_hat inherits from `NULL` not `numeric`.
    
     ── 3. Error: parsnip (@test_yhat_model_info.R#168) ────────────────────────────
     argument is of length zero
     Backtrace:
     1. testthat::expect_length(...)
     4. DALEX:::yhat.model_fit(...)
     7. parsnip::predict.model_fit(X.model, newdata, type = "prob")
     9. parsnip::predict_classprob.model_fit(...)
     10. parsnip:::prepare_data(object, new_data)
     11. parsnip:::convert_form_to_xy_new(object$preproc, new_data)
    
     Wrapper package: caret
     Wrapper package version: 6.0.86
     Package: randomForest
     Package version: 4.6.14
     Task type: classification══ testthat results ═══════════════════════════════════════════════════════════
     [ OK: 174 | SKIPPED: 0 | WARNINGS: 6 | FAILED: 3 ]
     1. Failure: parsnip (@test_yhat_model_info.R#164)
     2. Failure: parsnip (@test_yhat_model_info.R#166)
     3. Error: parsnip (@test_yhat_model_info.R#168)
    
     Error: testthat unit tests failed
     Execution halted
Flavor: r-devel-linux-x86_64-fedora-gcc