Files
quantconnect--lean/Indicators/PythonIndicator.cs
T
Ricardo Andrés Marino Rojas fc6ddc2120
Build & Test Lean / build (push) Has been cancelled
Regression Tests / build (push) Has been cancelled
Feature 5988 WarmUpIndicator() method for indicators written in Python (#6027)
* Implement IIndicatorWarmUpPeriodProvider
- Implement IIndicatorWarmUpPeriodProvider in PythonIndicator.cs
- Make a unit test to check whether the WarmUpPeriod is working as expected
- Make a regression test to check the new feature at a system level

* Nit change

* Change Period parameter for WarmUpPeriod parameter
- Change regression test to check if the new parameter keep backwards compatibility with indicators that do not set WarmUpPeriod

* Documentation change

* Fix tests bugs
- In CommonIndicatorTests.cs before finish the test it checks the period.value with the number of samples but for default the period.value was set to -1

* Change names

* Change WarmUp and RegisterIndicator methods
- Lean WarmUp indicator skip custom python indicators that don't define WarmUpPeriod parameter

* Call WarmUpIndicator manually
- Add a new "bridge" method called WarmUpIndicator in QCAlgorithm.Python.cs to set up everything to call WarmUpIndicator in QCAlgorithm.Indicators.cs
- Change the regression algorithm to warm up the indicators manually

* Remove unnecessary code and add more tests

* Nit change

* Revert "Nit change"

This reverts commit da411f59c9e4295d75a11c6c581f615a938dd2c6.

* Fix bugs

* Try fix bugs

* Add C# regression test
- More nit changes
- Fix bugs

* Requested changes

* Remove unnecessary code

* Requested changes

* Nit changes
- Add new Python class to check a custom indicator, which doesn't inherits from PythonIndicator, warms up properly

* Reduce redundant code

* Fix bug and add more unit and regression tests

* - Add more unit tests

* Nit change

* Test cleanup

Co-authored-by: Martin-Molinero <martin@quantconnect.com>
2021-10-28 20:29:22 -03:00

149 lines
5.6 KiB
C#

/*
* QUANTCONNECT.COM - Democratizing Finance, Empowering Individuals.
* Lean Algorithmic Trading Engine v2.0. Copyright 2014 QuantConnect Corporation.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
using Python.Runtime;
using QuantConnect.Data;
using System;
namespace QuantConnect.Indicators
{
/// <summary>
/// Provides a wrapper for <see cref="IndicatorBase{IBaseData}"/> implementations written in python
/// </summary>
public class PythonIndicator : IndicatorBase<IBaseData>, IIndicatorWarmUpPeriodProvider
{
private bool _isReady;
private dynamic _indicator;
/// <summary>
/// Get the indicator Name. If not defined, use the class name
/// </summary>
/// <param name="indicator">The python implementation of <see cref="IndicatorBase{IBaseDataBar}"/></param>
/// <returns>The indicator Name.</returns>
private static string GetIndicatorName(PyObject indicator)
{
using (Py.GIL())
{
var name = indicator.HasAttr("Name")
? indicator.GetAttr("Name")
: indicator.GetAttr("__class__").GetAttr("__name__");
return name.GetAndDispose<string>();
}
}
/// <summary>
/// Get the indicator WarmUpPeriod parameter. If not defined, use 0
/// </summary>
/// <param name="indicator">The python implementation of <see cref="IndicatorBase{IBaseDataBar}"/></param>
/// <returns>The WarmUpPeriod of the indicator.</returns>
private static int GetIndicatorWarmUpPeriod(PyObject indicator)
{
using (Py.GIL())
{
var warmUpPeriod = indicator.HasAttr("WarmUpPeriod")
? indicator.GetAttr("WarmUpPeriod")
: 0.ToPython();
return warmUpPeriod.GetAndDispose<int>();
}
}
/// <summary>
/// Initializes a new instance of the PythonIndicator class using the specified name.
/// </summary>
/// <remarks>This overload allows inheritance for python classes with no arguments</remarks>
public PythonIndicator()
: base("")
{
}
/// <summary>
/// Initializes a new instance of the PythonIndicator class using the specified name.
/// </summary>
/// <remarks>This overload allows inheritance for python classes with multiple arguments</remarks>
public PythonIndicator(params PyObject[] args)
: base (GetIndicatorName(args[0]))
{
}
/// <summary>
/// Initializes a new instance of the PythonIndicator class using the specified name.
/// </summary>
/// <param name="indicator">The python implementation of <see cref="IndicatorBase{IBaseDataBar}"/></param>
public PythonIndicator(PyObject indicator)
: base (GetIndicatorName(indicator))
{
SetIndicator(indicator);
}
/// <summary>
/// Sets the python implementation of the indicator
/// </summary>
/// <param name="indicator">The python implementation of <see cref="IndicatorBase{IBaseDataBar}"/></param>
public void SetIndicator(PyObject indicator)
{
using (Py.GIL())
{
foreach (var attributeName in new[] {"IsReady", "Update", "Value"})
{
if (!indicator.HasAttr(attributeName))
{
var name = indicator.GetAttr("__class__").GetAttr("__name__");
var message = $"Indicator.{attributeName} must be implemented. " +
$"Please implement this missing method in {name}";
if (attributeName == "IsReady")
{
message += " or use PythonIndicator as base:" +
$"{Environment.NewLine}class {name}(PythonIndicator):";
}
throw new NotImplementedException(message);
}
}
}
WarmUpPeriod = GetIndicatorWarmUpPeriod(indicator);
_indicator = indicator;
}
/// <summary>
/// Gets a flag indicating when this indicator is ready and fully initialized
/// </summary>
public override bool IsReady => _isReady;
/// <summary>
/// Required period, in data points, for the indicator to be ready and fully initialized
/// </summary>
public int WarmUpPeriod { get; protected set; }
/// <summary>
/// Computes the next value of this indicator from the given state
/// </summary>
/// <param name="input">The input given to the indicator</param>
/// <returns>A new value for this indicator</returns>
protected override decimal ComputeNextValue(IBaseData input)
{
using (Py.GIL())
{
_isReady = _indicator.Update(input) ?? _indicator.IsReady;
return _indicator.Value;
}
}
}
}