Lots of messages printed out when I evaluate the model using MLJ

[goerch]

@Sijun

using MLJ

X, y = @load_iris

LogisticClassifier = @load LogisticClassifier pkg=MLJLinearModels verbosity=0

lc = LogisticClassifier()

evaluate(lc, X, y,
        resampling=CV(shuffle=true),
        measures=[log_loss, accuracy],
        verbosity=0)

Here are the results of a run on my system with Julia (nightly)

PerformanceEvaluation object with these fields:
  measure, measurement, operation, per_fold,
  per_observation, fitted_params_per_fold,
  report_per_fold, train_test_pairs
Extract:
┌────────────────────────────┬─────────────┬──────────────┬────────────────────────────────────────────┐
│ measure                    │ measurement │ operation    │ per_fold                                   │
├────────────────────────────┼─────────────┼──────────────┼────────────────────────────────────────────┤
│ LogLoss(tol = 2.22045e-16) │ 0.69        │ predict      │ [0.699, 0.679, 0.703, 0.732, 0.684, 0.646] │
│ Accuracy()                 │ 0.847       │ predict_mode │ [0.92, 0.88, 0.84, 0.84, 0.72, 0.88]       │
└────────────────────────────┴─────────────┴──────────────┴────────────────────────────────────────────┘

The messages you are seeing indicate internal compiler errors (and normally that is pretty serious). I’d suspect problems with your setup (Julia version/Package versions).