def dataListener(dataQueue, resQueue, gpuNum=None):
# Environment = env_utils.getEnvironment('cube3')
nnet = nnet_utils.loadNnet('savedModels/cube3/1/', 'model.meta"', True, Environment, gpuNum=gpuNum)
while True:
data = dataQueue.get()
nnetResult = nnet(data)
resQueue.put(nnetResult)
def dataListener(dataQueue, resQueue, args, useGPU, Environment, gpuNum=None):
nnet = nnet_utils.loadNnet(args.model_loc,
args.model_name,
useGPU,
Environment,
gpuNum=gpuNum)
while True:
data = dataQueue.get()
nnetResult = nnet(data)
resQueue.put(nnetResult)
def dataListener(dataQueue, resQueue, gpuNum=None):
# 导入神经网络模型
nnet = nnet_utils.loadNnet(model_loc,
model_name,
False,
Environment,
gpuNum=gpuNum)
while True:
data = dataQueue.get()
nnetResult = nnet(data)
resQueue.put(nnetResult)
def heurProc(dataQueue,resQueue,modelLoc,useGPU,Environment,gpuNum=None):
if modelLoc == "":
nnet = lambda x,realWorld: np.zeros([x.shape[0]],dtype=np.int)
else:
nnet = nnet_utils.loadNnet(modelLoc,"",useGPU,Environment,gpuNum=gpuNum)
while True:
data, realWorld = dataQueue.get()
if data is None:
resQueue.put(None)
break
nnetResult = nnet(data,realWorld=realWorld)
resQueue.put(nnetResult)
def dataListener(dataQueue, resQueue, gpuNum=None):
model_loc = os.path.join(BASE_DIR, "DeepCube/savedModels/cube3/1/")
model_name = 'model.meta'
Environment = env_utils.getEnvironment('CUBE3')
#nnet = nnet_utils.loadNnet(model_loc,model_name,useGPU,Environment,gpuNum=gpuNum)
nnet = nnet_utils.loadNnet(model_loc,
model_name,
True,
Environment,
gpuNum=gpuNum)
while True:
data = dataQueue.get()
nnetResult = nnet(data)
resQueue.put(nnetResult)
modelLoc = args.model_loc
numStates = args.num_states
maxTurns = args.max_turns
searchDepth = args.search_depth
numRollouts = args.num_rollouts
solveMethod = args.method.upper()
verbose = args.verbose
if maxTurns is None:
maxTurns = args.max_s
load_start_time = time.time()
if solveMethod == "BFS" or solveMethod == "MCTS" or solveMethod == "MCTS_SOLVE" or solveMethod == "BESTFS":
assert (modelLoc != "")
### Restore session
heuristicFn = nnet_utils.loadNnet(args.model_loc, args.model_name,
not args.noGPU, Environment)
print("Loaded: %s" % (time.time() - load_start_time))
### Run network on different scrambles
scrambleTests = range(args.min_s, args.max_s, 1)
if args.max_s - args.min_s > 30:
scrambleTests = np.linspace(args.min_s, args.max_s, 30, dtype=np.int)
for scrambleNum in scrambleTests:
solve_start_time = time.time()
# Solve cubes
testStates_cube, _ = Environment.generate_envs(numStates,
[scrambleNum, scrambleNum])
testStates = Environment.state_to_nnet_input(np.stack(testStates_cube))
if solveMethod == "BFS" or solveMethod == "MCTS":
for move in moves:
self.state = self.env.next_state(self.state, move)
self._updatePlot()
time.sleep(0.5)
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--N', type=int, default=5, help="")
parser.add_argument('--heur', type=str, default=None, help="")
parser.add_argument('--init', type=str, default=None, help="")
args = parser.parse_args()
env = LightsOut(args.N)
if args.init is None:
state = np.array([env.solvedState])
else:
state = np.array(
[[int(x) for x in list(re.sub("[^0-9]", "", args.init))]])
#state = np.array(env.generate_envs(1, [100, 100])[0])
heuristicFn = None
if args.heur is not None:
heuristicFn = nnet_utils.loadNnet(args.heur, "", False, env)
fig = plt.figure(figsize=(5, 5))
interactiveEnv = InteractiveEnv(state, env, heuristicFn)
fig.add_axes(interactiveEnv)
plt.show()