I’m working with PyCommunicator in AnyLogic to run a .py script and retrieve its output. For basic Python commands like print() or preprocessing steps, the output appears correctly in the AnyLogic console. However, when I execute a Machine Learning block (e.g., model training using scikit-learn or XGBoost), the script seems to run (judging by CPU activity), but no output is displayed in AnyLogic — neither during execution nor after completion. I’ve tested the same script in Spyder, and it runs fine within a few minutes, producing all expected results. The “Redirect Python Output” option is already checked in the PyCommunicator settings. Why does PyCommunicator fail to display output when the Python script includes ML training and how to resolve this issue? Continue reading...
