Backtesting for scanner class

Hi Dharmesh,

QuantConnect always runs main.py as the entry point for backtests and does not allow you to select another file (like scanner.py) to run directly. However, you can easily use your scanning logic defined in scanner.py by following one of these approaches:

1. Copy into main.py:
Copy the relevant class, function, or logic directly from scanner.py into main.py.

2. Import in main.py:

• Leave your scanner logic in scanner.py.
• In your main.py, add import scanner.
• Use your scanner functionality by creating an instance or calling the method you need, such as scanner.DataScanner(self).scan_data() within your QCAlgorithm subclass.

Note: Your main.py must contain the subclass of QCAlgorithm and its initialize method—QuantConnect requires this.

Example (option 2: importing from scanner.py):

If scanner.py contains:

class DataScanner:
    def __init__(self, algo):
        self._algo = algo
    def scan_data(self):
        # scanning logic here
        self._algo.debug("Scan started")

Then, in main.py:

from AlgorithmImports import *
import scanner

class MyAlgorithm(QCAlgorithm):
    def initialize(self):
        self.set_start_date(2021, 1, 1)
        self.set_end_date(2022, 1, 1)
        self.set_cash(100_000)

        self._scanner = scanner.data_scanner(self)
        self._scanner.scan_data()  # Example call at startup

    def on_data(self, data):
        pass

You can also schedule or trigger the scan from within main.py as needed.

Summary: QuantConnect will always run main.py first, but you can re-use and modularize your code by importing your scanner functions or classes into main.py.

Let me know if you want to see more detailed examples or need help adapting your code!

I have so much logic so perhaps I want to short circuit all other initialization in main.py and only run scanner method ?