Excellent tutorial series. Obviously a huge amount of work has gone into coding, building and testing models to generate positive results. Also a lot of component modules which are innovative, not really published elsewhere insights.
Thank you for sharing. This is a great content. I followed your steps and made several changes to meet what I need. I finally got what I want after studying for 4 months and of-cuz thanks to your videos. Remaining task for is to deploy it on real market. My question to you is do I need to create a new environment to reflect the actual trading environment?
I would like to suggest that could improved the performance on the ML is inserting a higher time frame(TF) indicator for it to identify the trend. e.g if we are on a H1 TF, we would add a 4H TF to it. This should make the bot avoid trading against the current trend and make the trade count lessor. I think higher time frame could be done with pandas resample method, hope to hear from you soon, and take care. Thanks for the awesome code Rokas Balsys.
Thanks for this great tutorial. Deep Reinforced Learning is amazing… Quick Question: How do I retrain a model so I dont have to start from scratch every time? ie how would i load the actor and critic model to retrain?
wow! I was in the middle of testing correlations to get to that 30 something indicator level but you just answered most the questions here with a live example.. Thank you again, you're awesome!
yes, very good stuff about DRL, it's an other virtual trading with training and testing in the desktop, I think python is made also for real trading, and until now ther is no article in the internet which handle how to implement these concepts in a real trading.
Great work man, lets all work together to develop something and get rich together ! Would appreciate a video to implement this on Binanace exchange using their APIs.
I have almost finished the bot connected to binance, if anyone is interested in exchanging knowledge to finish connecting it correctly without errors, I am at your disposal!
you really motivated me to learn machine learning, thank you so much. Hope you still continue your passion, and get 100k Sub soon ^^ From Vietnam With Love
Very Nice tutorial, I have a question, After the training, Hot much time the model will work? I mean, is better to recreate the model after a week for example? Thank you!
Thank you so so much for this.
Has anyone able to refine or make changes to the code and improve it, great as it is, if so would you be so kind and share?
Thanks a lot for sharing this tutorial. I am having issues where the model wont train.. has anyone else had this issue?
Excellent tutorial series. Obviously a huge amount of work has gone into coding, building and testing models to generate positive results. Also a lot of component modules which are innovative, not really published elsewhere insights.
Thank you for sharing. This is a great content. I followed your steps and made several changes to meet what I need. I finally got what I want after studying for 4 months and of-cuz thanks to your videos. Remaining task for is to deploy it on real market. My question to you is do I need to create a new environment to reflect the actual trading environment?
please make a tutorial how we connect to exchange api demo or real account!!
Thank you very much. You made us learn a lot of things. Really thank you.
I would like to suggest that could improved the performance on the ML is inserting a higher time frame(TF) indicator for it to identify the trend. e.g if we are on a H1 TF, we would add a 4H TF to it. This should make the bot avoid trading against the current trend and make the trade count lessor. I think higher time frame could be done with pandas resample method, hope to hear from you soon, and take care.
Thanks for the awesome code Rokas Balsys.
very great stuff. Thanks a lot!
Thanks for this great tutorial. Deep Reinforced Learning is amazing… Quick Question: How do I retrain a model so I dont have to start from scratch every time? ie how would i load the actor and critic model to retrain?
wow! I was in the middle of testing correlations to get to that 30 something indicator level but you just answered most the questions here with a live example.. Thank you again, you're awesome!
This is great stuff. Cloned your repo, configuring and testing your env… will make sure to credit your work. Thank you.
we just want for the author to make an other video and show us how he can connect this bot to binance exchange or using MQ5 api.
yes, very good stuff about DRL, it's an other virtual trading with training and testing in the desktop, I think python is made also for real trading, and until now ther is no article in the internet which handle how to implement these concepts in a real trading.
How to save model and retrain in parts ??
Great work man, lets all work together to develop something and get rich together !
Would appreciate a video to implement this on Binanace exchange using their APIs.
I have almost finished the bot connected to binance, if anyone is interested in exchanging knowledge to finish connecting it correctly without errors, I am at your disposal!
you really motivated me to learn machine learning, thank you so much. Hope you still continue your passion, and get 100k Sub soon ^^ From Vietnam With Love
would it be possible to connect this bot per api to live data from an exchange after training it?
Perfect, How can i reach you ?
this is amazing
Very Nice tutorial, I have a question, After the training, Hot much time the model will work? I mean, is better to recreate the model after a week for example? Thank you!