I want to use the Python versions of TabPFN and AutoGluon via the reticulate package in R. When I run the code on my laptop (Windows), everything works fine. However, as soon as I run the same code on my Mac Mini, the R session crashes. With AutoGluon, it usually happens during prediction, and with TabPFN, it typically occurs when the module is being loaded. (And I really want to use R—for various reasons.) Maybe someone has experienced similar issues when using macOS vs. Windows and can help me out. This is the used code: library(mlr3) library(mlr3tuning) library(reticulate) # tabpfn tabpfn = import('tabpfn') set.seed(42) n = 30 p = 5 X = matrix(rnorm(n * p), ncol = p) y = sample(0:1, n, replace = TRUE) py_X = r_to_py(X) py_y = r_to_py(y) tabpfn_model = tabpfn$TabPFNClassifier(device = 'cpu') tabpfn_model$fit(py_X, py_y) predictions = tabpfn_model$predict(py_X) r_predictions = py_to_r(predictions) print(r_predictions) # autogloun autogluon = import('autogluon.tabular') train_data = cbind(X, target = y) py_train_data = r_to_py(train_data) predictor = autogluon$TabularPredictor(label = 'target') predictor$fit( train_data = py_train_data, presets = 'best_quality', time_limit = 30 ) predictions = predictor$predict(py_train_data) r_predictions = py_to_r(predictions) print(r_predictions) Additional some system information: py_config() python: /Users/Dummy/.virtualenvs/r-reticulate/bin/python libpython: /Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/config-3.12-darwin/libpython3.12.dylib pythonhome: /Users/Dummy/.virtualenvs/r-reticulate:/Users/Dummy/.virtualenvs/r-reticulate version: 3.12.0 (v3.12.0:0fb18b02c8, Oct 2 2023, 09:45:56) [Clang 13.0.0 (clang-1300.0.29.30)] numpy: /Users/Dummy/.virtualenvs/r-reticulate/lib/python3.12/site-packages/numpy numpy_version: 2.1.3 Thank you very much Continue reading...