如何使用subprocess调用Python脚本并获取函数返回值Y_pred

Got it, let's break down how to use the subprocess module to call your calculate.py script and capture the return value of mytest() into the returned_val variable.

Step 1: Prepare calculate.py for Output #

First, you need to make your script print the result of mytest() when it's run directly—since subprocess captures a script's standard output, not direct function return values. Using JSON serialization is a safe way to handle complex return types (like lists, dictionaries) without messy parsing:

# calculate.py
def mytest():
    # Replace this with your actual function logic
    return [10, 20, 30]  # Example return value

if __name__ == "__main__":
    import json
    # Serialize the result to JSON and print it
    print(json.dumps(mytest()))

Step 2: Use subprocess to Call the Script and Parse Output #

In your Python console or main script, use subprocess.run() to execute calculate.py, capture its output, and convert it back into a Python variable:

import subprocess
import json

try:
    # Run the script and capture stdout/stderr
    proc_result = subprocess.run(
        ["python", "calculate.py"],  # Use "python3" instead if needed on your system
        capture_output=True,
        text=True,  # Return output as string instead of raw bytes
        check=True  # Raises an exception if the script exits with an error code
    )
    
    # Parse the JSON output into a Python object
    returned_val = json.loads(proc_result.stdout.strip())
    print(f"Captured value: {returned_val}")
except subprocess.CalledProcessError as e:
    print(f"Error running script: {e.stderr}")
except json.JSONDecodeError as e:
    print(f"Failed to parse output: {e}")

Quick Tips: #

• Path Handling: If calculate.py isn't in your current working directory, replace "calculate.py" with its full absolute path (e.g., "/home/user/projects/calculate.py").
• Python Executable: On some systems, Python 3 is called with python3 instead of python—adjust the command list accordingly.
• Simple Return Types: If mytest() returns a basic type like an integer or string, you can skip JSON and directly cast the output (e.g., returned_val = int(proc_result.stdout.strip())), but JSON is more robust for complex data.
• Error Handling: The try/except blocks help catch issues like script crashes or invalid output formatting.

内容的提问来源于stack exchange，提问作者user1670773