Currently the application calls for a hard-coded python environment. My fix is to use sys.executable to dynamically launch GUI instead of hard-coded env path. Currently, main.py launches the GUI by calling a fixed Python executable path under: ./envs/env-base/python.exe. This causes failures when AddaxAI is installed in a different location, when a user relies on a different virtual environment, or when running within Conda.
Replacing the call to get_python_interprator("base") with: subprocess.run([sys.executable, GUI_script]) fixes the issue. Additionally, deprecating the get_python_interprator() function entirely in both main.py and Addax_GUI.py streamlines the fix.