PyTorch-DDPG-Stock-Trading
PyTorch-DDPG-Stock-Trading copied to clipboard
An implementation of DDPG using PyTorch for algorithmic trading on Chinese SH50 stock market.
I think below code maybe has problem: https://github.com/JoshuaWu1997/PyTorch-DDPG-Stock-Trading/blob/0c5c2f9095f5871d26b573fd960fb772f4ea050b/DDPG.py#L85-L90 The two gradients need to be calculated separately, because their loss functions are different. I think it should be changed to below:...
below code: https://github.com/JoshuaWu1997/PyTorch-DDPG-Stock-Trading/blob/0c5c2f9095f5871d26b573fd960fb772f4ea050b/DDPG.py#L120 Shouldn't above line of code be modified to below? `allocation = self.actor_network.actor_action(next_state_data).data.view(-1)` thanks!