I am working to build an deep reinforcement learning agent which can place orders (i.e. limit buy and limit sell orders). The actions are {"Buy": 0 , "Do Nothing": 1, "Sell": 2}.