Deep-Reinforcement-Learning-in-Trading icon indicating copy to clipboard operation
Deep-Reinforcement-Learning-in-Trading copied to clipboard

Published 20 hours ago • verified publisher icon saeed349

test_codes for dqn and ddqn

Open parthvadhadiya opened this issue 7 years ago • 3 comments

hey. wonderful work sir. i have read your report also. you mentioned about dqn and Double DQN. if you don't mind can i also look into those implementation? just want to see clear picture of the loopholes with those and also want to take this project to further advancements. like as you mentioned , Prioritized Experience Replay.

parthvadhadiya avatar Nov 14 '18 11:11 parthvadhadiya

Hi Parth, Glad you liked it. Code for Dqn, double and duelling Dqn are all in the subfolder Agent

saeed349 avatar Nov 14 '18 12:11 saeed349

my apologize, thank you for kindly response.

parthvadhadiya avatar Nov 14 '18 12:11 parthvadhadiya

in middle of the sea i show there is stock splitting at some stages on mostly every instrument(stock). did you encounter this or just leave it as it is? worrying it destruct my model.

parthvadhadiya avatar Dec 05 '18 09:12 parthvadhadiya