Back

Deep Learning in QuantConnect

Hey all, I have been live trading deep learning algos on another platform for a while now but have recently been lured here by the tick granularity quant connect offers.. I just wanted to get your experience in a couple things:
(1) how do the deep learning libraries compare in C#? Python/R are indisputably the forerunners if you wish to implement ML and I would need something comparable to make to switch.
(2)How easy is it to set up live trading with IB? I know it will be a bit a work but I just want to be assured there aren't any outstanding bugs/issues with implementing this
(3) any other cautionary comments?
PS thanks ahead of time, I'm just trying to make sure I'll have what I need before investing a large amount of time here!
Update Backtest








Welcome to QuantConnect @Ethn, we have two AI libraries white listed though I can't be sure they'll do what you need. I'd recommend looking at them - AForge and Accord.NET. We've also already finished and tested the IB trading code - its all open sourced on Github.

I suppose time required for porting your algorithm would be dependent on how many core python functions/behaviors you're using. But once you've made the switch I think you'll start enjoying type safety.. it helps a lot in the long run :)
0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


For R code using is the best way to use R.NET from NuGet.
1

It's hard to say exactly without knowing what features you're using from the python ml library.

I've had great success using AForge.NET's genetic library. You can check out their nuero and machine learning libraries:
http://www.aforgenet.com/framework/docs/

If you can let me know what features you're looking for I can let you know if they're supported or how hard it would be to add those features to those open sourced libraries.
0

Thanks for pointing me there michael, it looks as if there is functionality for hidden layers in neural nets which was the major thing I was concerned about... Many libraries only use single layer perceptron learning which isn't sufficient for price/technicals patterns.
0

Update Backtest





0

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


Loading...

This discussion is closed