Suppose we already have a python package/function created on a local machine. This article will guide you through the steps to plug into ALGOGENE for backtest, live-test or real-trading.

Preparation

For example, we created a simple package called "my_model" on a desktop machine.

The whole package just has an empty "__init__.py" file, with a module file "func1.py". The module file contains 3 simple functions as follows.

File Upload

Then, we can upload the package to ALGOGENE as follows.

• after login the platform, go to [My History] > [Custom File Viewer]
• select "/lib" in the cloud directory, then drag our local packge on it


Figure 3 - upload files to ALGOGENE cloud

Upon successful upload, the cloud directory should look something like.

Backtest Example

Suppose we want to use "my_diff" function in "func1.py" to calculate the market spread and print the result to console. It can be done as follows:

• after "self.evt" is initialized (at line 8)
• use "self.evt._include(...)" to include our package in python runtime environment (at line 11)
• import our desired module, and declare the module as global variable (at line 12-13)
• we can then call the module functions anywhere in the backtest script (at line 19)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20

from AlgoAPI import AlgoAPIUtil, AlgoAPI_Backtest

class AlgoEvent:
    def __init__(self):
        pass
        
    def start(self, mEvt):
        self.evt = AlgoAPI_Backtest.AlgoEvtHandler(self, mEvt)

        # include my package to python environment, then import a function
        self.evt._include("my_model")
        global func1
        import func1
        
        self.evt.start()
        
    def on_marketdatafeed(self, md, ab):
        # call my function
        v = func1.my_diff(md.askPrice, md.bidPrice)
        self.evt.consoleLog(v)

We can find the calculated results from console.

That's it!