Unleashing Spectacular Powers Of AI Trading With More Improvements (FinRock)

Welcome back to another video where we delve into the intricacies of live trading on test data. Join me as I navigate the developments and changes in my trading model, utilizing reinforcement learning to execute buy and sell orders. While the data showcased here is simulated, it is a pivotal tool in tracking the performance and progress of my trading algorithms.

In this video, I'll walk you through the latest updates and enhancements to my trading framework. Every aspect has been meticulously refined to optimize performance and reliability, from revamped indicators to improved reward functions and enhanced scalers. Additionally, I'll demonstrate how configurations are now seamlessly integrated, allowing for more straightforward adaptation and replication of strategies.

As we explore the code and its functionalities, you'll gain insights into the continuous evolution of this project. From discussing the rationale behind each modification to showcasing the results of extensive training, this video offers a comprehensive overview of the advancements achieved.

Whether you're a seasoned trader looking to fine-tune your strategies or a budding enthusiast eager to explore algorithmic trading, this tutorial provides valuable insights and practical guidance. So please sit back, engage with the content, and let's embark on this journey of innovation and discovery in algorithmic trading.

Tutorial part #5: This Video

#FinRock #AI #ReinforcementLearning #TradingIndicators #MetricsMagic #Python

[gordongekko]

Thanks for the update. I'm having trouble running it, simplereward uppercase? Needs to be instantiated first?

[amoose]

how do you open the tensor board results for testing set?
cant seem to find the logs for that

[ahmedmidany]

how can we train it with real historical data for specific coin and then feed in realtime live data?

[capitola]

very nice. Do you have a list of future features and the time plan? Like adding shorts, adding stop losses and/or take profits learned or configurable such as: SL=1xATR TP=risk * reward?

[amoose]

what does the different actions represent when printing results?
does the number represent the total number of actions within that step?

[FreeVideoGames_]

Cool software 😎 is this in tradingview or MT4 or MT5?

[mls]

Is it an open project? We are waiting for new features.

[МишаЦобан]

Ты просто гений. Я хочу научиться писать код также как ты. Молодец

[vasiliychernenko]

can you explain this math?
if action == 2: # buy
buy_order_size = order_size
= + (1 - * buy_order_size
next_state.assets = last_state.assets + (last_state.balance * buy_order_size / last_state.close) * self.fee_ratio
next_state.balance = last_state.balance - (last_state.balance * buy_order_size) * self.fee_ratio

elif action == 1: # sell
sell_order_size = order_size
= - * sell_order_size
next_state.balance = last_state.balance + (last_state.assets * sell_order_size * last_state.close) * self.fee_ratio
next_state.assets = last_state.assets - (last_state.assets * sell_order_size) * self.fee_ratio