ฝัง
- เผยแพร่เมื่อ 7 ก.ย. 2024
- Hey, thanks for clicking on the video.
JOIN THE NEWSLETTER
witty-motivato...
I talk about coding, python, technology, education, data science, deep learning, etc. If you enjoyed the video please like or subscribe. It is one of the best ways to let TH-cam share similar content to you and others interested in this topic.Many thanks
GET THE CODE
colab.research...
CHAPTERS
Portfolio - 4:00
Process Data - 5:19
Split Data - 7:20
Save Data - 7:50
My goal is to create a community of like-minded people for a mastermind group where we can help each other succeed, so browse around and let me know what you think. Cheers!
Keyword for the algorithm: creating a stock trading ai agent with FinRL reinforcement learning deep learning finance $aapl data science python coding quant market artificial intelligence neural network
FinRL is a must! 👏 Good job!
Yes! Thank you!
Fantastic Work! Did not know this process was called Wrangling
Thanks for watching-- that's what I was taught, but could totally be wrong too. I've also heard it referred to as data gathering or data cleaning.
AWESOME
Thank you! Cheers!
Although the concept is very promising, but it is broken. If you check the actions of each agent, you will observe that the agents buy few stocks on the first trading day and holds on to them throughout the trading period. The agents did not sell those stocks even when their were sharp fall in price. It is not clear to me how to improve the performance of those agents. In the documentation also it is not mentioned. Hyper-parameter tuning/ Optimization might help which i did not try.
Sounds like you have a bit more experience with FinRL than I do. I still haven't produced a backtest that significantly outperforms benchmarks. I'm guessing some hyper-tuning or even a custom rewards function could help make better decisions by the agent.
@@eminshI am not expert in FinRL but an enthusiast like you. I had huge expectation and when i got poor result, I was highly disappointed. In the documentation they have only talked about hypermarameter tuning optimization to improve the agents decision making. I guess they have not provided the optimized configurations for the agents in the examples. We need more knowledge regarding RL to improve the performance of agents.
Hello. It seems this code doesn't work anymore importing yahoo library keeps throwing error and I also can't find from installation from FinRL git
the problem is still there
Hi eminshall, can you share the notebook or git
I will post notebook to my GitHub
Do you plan to use elegantrl or stablebaselines when training the agent? I've been having trouble dealing with dependency issues between the different packages.
I used stablebaselines, but I'll try elegant too.
i got error on this line :
from finrl.meta.preprocessor.preprocessors import FeatureEngineer, data_split
i look into it but no result