cvigoe

NUM_ALGO_TRIALS = _
NUM_TRAINING_TIMESTEPS = _
NUM_TESTING_TIMESTEPS = _

set_seed()
for algo_iteration in range(NUM_ALGO_TRIALS):
	for training_timesteps in range(NUM_TRAINING_TIMESTEPS): #for epsiodic tasks, keep resetting the env
		deploy_policy_with_exploration() 		
		update_policy_with_new_data()
		for testing_timesteps in range(NUM_TESTING_TIMESTEPS): #for epsiodic tasks, keep resetting the env

cvigoe

NUM_ALGO_TRIALS = _
NUM_TRAINING_TIMESTEPS = _
NUM_TESTING_TIMESTEPS = _

set_seed()
for algo_iteration in range(NUM_ALGO_TRIALS):
	for training_timesteps in range(NUM_TRAINING_TIMESTEPS): #for epsiodic tasks, keep resetting the env
		deploy_policy_with_exploration() 		
		update_policy_with_new_data()
		for testing_timesteps in range(NUM_TESTING_TIMESTEPS): #for epsiodic tasks, keep resetting the env

cvigoe

def simulator(initial_state, parameter):
	state = initial_state
	for t in range(timesteps):
		state = next_state(state,parameter)
		print(state)


#############################################
# but I need to be able to do something like

cvigoe

#######################################################################################################################
# Simulation of Bayesian Dwell Time Model Assuming Error-free Covariates
# Conor Igoe - 2017
#
# Loads relevant data from the test directory given by test_data_filepath
# Initialises the necessary data structures for PyMC3 and logging
# Steps through the loaded test data chronologically
# 	For the 1st data point, with no posteriors to use as priors for the model, uses uniform priors
# 	For all other data points, uses posteriors from the previous datapoint as priors for current point in simulation
# 	Calculates the likelihood