logging.basicConfig(level=logging.INFO)
# --- Parse parameters ---
parameters = process_args(sys.argv[1:], Defaults)
if parameters.deterministic:
rng = np.random.RandomState(123456)
else:
rng = np.random.RandomState()
# --- Instantiate environment ---
env = ALE_env(rng,
frame_skip=parameters.frame_skip,
ale_options=[{
"key": "random_seed",
"value": rng.randint(9999)
}, {
"key": "color_averaging",
"value": True
}, {
"key": "repeat_action_probability",
"value": 0.
}])
# --- Instantiate qnetwork ---
qnetwork = MyQNetwork(env, parameters.rms_decay, parameters.rms_epsilon,
parameters.momentum, parameters.clip_delta,
parameters.freeze_interval, parameters.batch_size,
parameters.update_rule, rng)
test_policy = EpsilonGreedyPolicy(qnetwork, env.nActions(), rng, 0.05)
# --- Instantiate agent ---
if __name__ == "__main__":
logging.basicConfig(level=logging.INFO)
# --- Parse parameters ---
parameters = process_args(sys.argv[1:], Defaults)
if parameters.deterministic:
rng = np.random.RandomState(123456)
else:
rng = np.random.RandomState()
# --- Instantiate environment ---
env = ALE_env(
rng,
frame_skip=parameters.frame_skip,
ale_options=[
{"key": "random_seed", "value": rng.randint(9999)},
{"key": "color_averaging", "value": True},
{"key": "repeat_action_probability", "value": 0.0},
],
)
# --- Instantiate qnetwork ---
qnetwork = MyQNetwork(
env,
parameters.rms_decay,
parameters.rms_epsilon,
parameters.momentum,
parameters.clip_delta,
parameters.freeze_interval,
parameters.batch_size,
parameters.update_rule,
logging.basicConfig(level=logging.INFO)
# --- Parse parameters ---
parameters = process_args(sys.argv[1:], Defaults)
if parameters.deterministic:
rng = np.random.RandomState(123456)
else:
rng = np.random.RandomState()
# --- Instantiate environment ---
env = ALE_env(rng,
frame_skip=parameters.frame_skip,
ale_options=[{
"key": "random_seed",
"value": rng.randint(9999)
}, {
"key": "color_averaging",
"value": True
}, {
"key": "repeat_action_probability",
"value": 0.
}])
# --- Instantiate qnetwork ---
qnetwork = MyQNetwork(env, parameters.rms_decay, parameters.rms_epsilon,
parameters.momentum, parameters.clip_delta,
parameters.freeze_interval, parameters.batch_size,
parameters.network_type, parameters.update_rule,
parameters.batch_accumulator, rng)
# --- Instantiate agent ---
agent = ALEAgent(
if __name__ == "__main__":
logging.basicConfig(level=logging.INFO)
# --- Parse parameters ---
parameters = process_args(sys.argv[1:], Defaults)
if parameters.deterministic:
rng = np.random.RandomState(123456)
else:
rng = np.random.RandomState()
# --- Instantiate environment ---
env = ALE_env(rng, frame_skip=parameters.frame_skip,
ale_options=[{"key": "random_seed", "value": rng.randint(9999)},
{"key": "color_averaging", "value": True},
{"key": "repeat_action_probability", "value": 0.}])
# --- Instantiate qnetwork ---
qnetwork = MyQNetwork(
env,
parameters.rms_decay,
parameters.rms_epsilon,
parameters.momentum,
parameters.clip_delta,
parameters.freeze_interval,
parameters.batch_size,
parameters.network_type,
parameters.update_rule,
parameters.batch_accumulator,
rng)