A Horrible mistake!

[kianfa]

I noticed that you are reseting the env at the beginning of each episode!

By doing this you never go forward in your dataset!

check train_agent function.

Thats why you train your model for 50000 episodes and nothing goes wrong!! (all your database is smaller than 50000. it is 23450)

[manigot]

does this affect multiprocess training though? i couldn't see any reset in multiprocess env.

[kianfa]

I didn't check that environment.
but your env at part 4 has this problem for sure.
I hardly remember what was wrong exactly. It was a long time ago. However, you might find something by searching "reset(" in your code. I believe you had called this function in your loop!

[kianfa]

by having this sentence in mind you might find the bug:

"I noticed that you are reseting the env at the beginning of each episode!"

[manigot]

Thanks. I'm not the repo owner though. I'm just a learner. By the way, could you suggest me some repos/tutorials for reinforcement or deep learning for cryptocurrency trading?

[kianfa]

not really. I do not.
this is a complex world!
I would continue this adventure for next few years. Maybe we can stay in touch. email me directly.

[pythonlessons]

Hello, this repo was not updated for 4 years, it is outdated and might be issues, bugs and etc. I recommend to switch to my later similar implementation: https://github.com/pythonlessons/FinRock