Applied Reinforcement Learning with Python: With OpenAI Gym, Tensorflow, and Keras

Appendix A: Source Code

This appendix references the initial release source code for this book. For updates that will happen occasionally to the code base as necessary, please check the Github by going to www.apress.com/9781484251263 .

Market Making Model Utilities

from collections import deque

class Memory():

    def __init__(self, max_size):

        self.buffer = deque(maxlen = max_size)

    def add(self, experience):

        self.buffer.append(experience)

    def sample(self, batch_size):

        buffer_size = len(self.buffer)

        index = np.random.choice(np.arange(buffer_size),

                                size=batch_size,

                                replace=True)

        return [self.buffer[i] for i in index]

class DeepQNetworkMM(): ...