Python Deep Learning Projects

Defining hyperparameters for Deep Q Learning (DQN)

The following are some of the hyperparameters defined that we will be using throughout the code and are totally configurable:

# Discount in Bellman Equationgamma = 0.95 # Epsilonepsilon = 1.0# Minimum Epsilonepsilon_min = 0.01# Decay multiplier for epsilonepsilon_decay = 0.99# Size of deque containerdeque_len = 20000# Average score needed over 100 epochstarget_score = 200# Number of gamesepisodes = 2000# Data points per episode used to train the agentbatch_size = 64# Optimizer for training the agentoptimizer = 'adam'# Loss for training the agentloss = 'mse'

The following are the parameters used:

• gamma : Discount parameter in the Bellman equation
• epsilon_decay: Multiplier by which you want ...