You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
hey, louisoutin
as we know, there is no holy grail in trading......
what the big diff between nlp&CV and live trading is: the market is changing its distribution, but the sentences or images not.
so I suggest you to do some walk forward test instead of traditional train, val, test.
Hey Diyano,
Thanks for the feedback/tips.
You are totally right, this is the whole challenge with trading data, I'll implement the walk forward test in my repo soon.
Do you thing that we have any chances to get profitable model with novel Attention/Transformers architectures?
I am currently developing a RL model for trading cryptocurrencies. For now, i'm at the first steps building a consistent approach, i.e. valuable reward function, kind of actions spaces, kinds of state spaces, etc.
Hey guys,
I did a fork of that project and made the following changes:
Here is the repo:
https://github.com/louisoutin/rat_crypto_trader
I didn't manage to get portfolio values greater than 1 for the moment (as this post mention too: #8).
The text was updated successfully, but these errors were encountered: