def transfer2JavaDblAry(gateway, pyArray, size):
dblAry = gateway.new_array(gateway.jvm.double, size)
i = 0
for x in pyArray:
dblAry[i] = float(x)
i = i + 1
return dblAry
def transfer2JavaStringAry(gateway, pyArray):
strAry = gateway.new_array(gateway.jvm.String, len(pyArray))
i = 0
for x in pyArray:
strAry[i] = str(x)
i = i + 1
return strAry
def _step(self, action):
assert self.action_space.contains(
action), "%r (%s) invalid" % (action, type(action))
# print(action)
# actionMapped = refer(action)
# print(actionMapped)
#actionPyAry = np.asarray(actionMapped,dtype = np.float64)
actionPyAry = np.asarray(action, dtype=np.float64)
# print(actionPyAry, 'len = ', actionPyAry.size)
# np array size = number of elements in the array
actionJavaAry = gateway.new_array(gateway.jvm.double, actionPyAry.size)
if (actionPyAry.size == 1):
actionJavaAry[0] = float(action)
else:
i = 0
for x in actionPyAry:
actionJavaAry[i] = x
i = i + 1
ipss_app.nextStepDynSim(self.step_time, actionJavaAry,
self.action_type)
# retrieve the state from InterPSS simulation service
# observations is a Java_Collections array
observations = ipss_app.getEnvironmentObversations()
# convert it from Java_collections array to native Python array
self.state = transfer2DJavaArray2NumpyArray(observations)
#check the states to see whether it go beyond the limits
done = ipss_app.isSimulationDone()
if not done:
reward = ipss_app.getReward()
elif self.steps_beyond_done is None:
self.steps_beyond_done = 0
reward = ipss_app.getReward(
) # even it is done, ipss_app would calculate and return a corresponding reward
else:
if self.steps_beyond_done == 0:
logger.warning(
"You are calling 'step()' even though this environment has already returned done = True. You should always call 'reset()' once you receive 'done = True' -- any further steps are undefined behavior."
)
self.steps_beyond_done += 1
reward = 0.0
return np.array(self.state).ravel(), reward, done, {}
