def mytest(env_name,
eval_episode=10,
num_init_traj=1,
max_horizon=15,
ensemble=1,
gt=0,
finetune=False,
finetune_iter=41,
finetune_proc=10,
cem_iter=20):
NUMS = {
'HalfCheetahPT-v2': 6,
'HopperPT-v2': 5,
'Walker2dPT-v2': 8,
}
num = NUMS[env_name]
if not finetune:
policy_net = get_awr_network(env_name, num)
else:
policy_net = get_finetune_network(env_name,
num,
num_iter=finetune_iter,
num_proc=finetune_proc)
model = make_parallel(10, env_name, num=num, stochastic=False)
env = make(env_name, num=num, resample_MP=True, stochastic=False)
params = get_params(env)
mean_params = np.array([0.5] * len(params))
osi = CEMOSI(model,
mean_params,
iter_num=cem_iter,
num_mutation=100,
num_elite=10,
std=0.3)
rewards, dist = online_osi(env,
osi,
policy_net,
num_init_traj=num_init_traj,
max_horizon=max_horizon,
eval_episodes=eval_episode,
use_state=False,
print_timestep=10000,
resample_MP=True,
ensemble=ensemble,
online=0,
gt=gt)
rewards = np.array(rewards)
print('l2 distance', dist)
print('rewards', rewards)
return {
'mean': rewards.mean(),
'std': rewards.std(),
'min': rewards.min(),
'max': rewards.max(),
'dist': dist.mean(),
}
def test_up_diff():
env_name = 'HopperPT-v2'
num = 5
policy_net = get_awr_network(env_name, num)
model = make_parallel(30, env_name, num=num, stochastic=False)
env = make(env_name, num=num, resample_MP=True, stochastic=False)
params = get_params(env)
#set_params(env, [0.55111654,0.55281674,0.46355396,0.84531834,0.58944066])
set_params(env,
[0.31851129, 0.93941556, 0.02147825, 0.43523052, 1.02611646])
set_params(env,
[0.94107358, 0.77519005, 0.44055224, 0.9369426, -0.03846457])
set_params(env,
[0.05039606, 0.14680257, 0.56502066, 0.25723492, 0.73810709])
mean_params = np.array([0.5] * len(params))
osi = DiffOSI(model, mean_params, 0.001, iter=100, momentum=0.9, eps=1e-3)
policy_net.set_params(mean_params)
# I run this at the last time..
# online is very useful ..
online_osi(env,
osi,
policy_net,
num_init_traj=5,
max_horizon=15,
eval_episodes=20,
use_state=False,
print_timestep=10000,
resample_MP=True,
online=0)
def test_UP():
env_name = 'DartHopperPT-v1'
num = 5
policy_net = get_up_network(env_name, num)
env = make(env_name, num=num, resample_MP=True)
eval_policy(policy_net,
env,
10,
0,
None,
timestep=1000,
use_state=False,
set_gt_params=True)
def test_up_osi():
#env_name = 'DartHopperPT-v1'
env_name = 'HopperPT-v2'
num = 5
#policy_net = get_up_network(env_name, num)
policy_net = get_awr_network(env_name, num)
model = make_parallel(10, env_name, num=num, stochastic=False)
env = make(env_name, num=num, resample_MP=True, stochastic=False)
params = get_params(env)
#set_params(env, [0.55111654,0.55281674,0.46355396,0.84531834,0.58944066])
set_params(env,
[0.31851129, 0.93941556, 0.02147825, 0.43523052, 1.02611646])
set_params(env,
[0.94107358, 0.77519005, 0.44055224, 0.9369426, -0.03846457])
set_params(env,
[0.05039606, 0.14680257, 0.56502066, 0.25723492, 0.73810709])
mean_params = np.array([0.5] * len(params))
osi = CEMOSI(model,
mean_params,
iter_num=20,
num_mutation=100,
num_elite=10,
std=0.3)
policy_net.set_params(mean_params)
online_osi(env,
osi,
policy_net,
num_init_traj=5,
max_horizon=15,
eval_episodes=30,
use_state=False,
print_timestep=10000,
resample_MP=True,
ensemble=1,
online=0,
gt=0)
def test_POLO():
#env_name = 'DartWalker2dPT-v1'
#num = 8
env_name = 'DartHopperPT-v1'
num = 5
value_net = get_td3_value(env_name)
#value_net = None
parser = argparse.ArgumentParser()
add_parser(parser)
args = parser.parse_args()
model = make_parallel(args.num_proc, env_name, num=num, stochastic=True)
env = make(env_name, num=num, resample_MP=False)
controller = POLO(value_net,
model,
action_space=env.action_space,
add_actions=args.add_actions,
horizon=args.horizon,
std=args.std,
iter_num=args.iter_num,
initial_iter=args.initial_iter,
num_mutation=args.num_mutation,
num_elite=args.num_elite,
alpha=0.1,
trunc_norm=True,
lower_bound=env.action_space.low,
upper_bound=env.action_space.high,
replan_period=5)
trajectories = eval_policy(controller,
env,
10,
args.video_num,
args.video_path,
timestep=args.timestep,
set_gt_params=True,
print_timestep=100)
import tensorflow as tf
import model
import data
# 訓練データ作成担当
g = data.Data()
# GPUをすべて使わないオプション
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
tf.keras.backend.set_session(sess)
# モデルを作成
model = model.make(tflite=False)
# 最適化を定義
optimizer = tf.keras.optimizers.Adam(lr=0.001)
model.compile(optimizer=optimizer,
loss="categorical_crossentropy",
metrics=["categorical_accuracy"])
# コールバック
class Callback(tf.keras.callbacks.Callback):
def on_epoch_end(self, epoch, logs=None):
"各エポック終了時に重みを保存する"
model.save("weight.hdf5")
cb = Callback()
# 途中から学習する場合
initial_epoch = 0
if initial_epoch >= 1:
def test_cem_osi():
env_name = 'HopperPT-v3'
num = 5
from networks import get_td3_value
#value_net = get_td3_value(env_name)
value_net = None
from policy import POLO, add_parser
import argparse
parser = argparse.ArgumentParser()
add_parser(parser)
args = parser.parse_args()
args.num_proc = 20
model = make_parallel(args.num_proc, env_name, num=num, stochastic=True)
env = make(env_name, num=num, resample_MP=True)
#args.iter_num = 2
args.num_mutation = 500
#args.num_mutation = 100
args.iter_num = 5
args.num_elite = 10
policy_net = POLO(value_net,
model,
action_space=env.action_space,
add_actions=args.add_actions,
horizon=args.horizon,
std=args.std,
iter_num=args.iter_num,
initial_iter=args.initial_iter,
num_mutation=args.num_mutation,
num_elite=args.num_elite,
alpha=0.1,
trunc_norm=True,
lower_bound=env.action_space.low,
upper_bound=env.action_space.high)
resample_MP = True
env = make(env_name, num=num, resample_MP=resample_MP, stochastic=False)
params = get_params(env)
print("FIXXXXXXXXXXXXXXXXXXXXXXPARAMETERS")
set_params(
env,
np.array([0.58093299, 0.05418986, 0.93399553, 0.1678795, 1.04150952]))
set_params(env,
[0.55111654, 0.55281674, 0.46355396, 0.84531834, 0.58944066])
set_params(env,
[0.31851129, 0.93941556, 0.02147825, 0.43523052, 1.02611646])
set_params(env, [0.58589476, 0.11078934, 0.348238, 0.68130195, 0.98376274])
mean_params = np.array([0.5] * len(params))
osi = CEMOSI(model,
mean_params,
iter_num=20,
num_mutation=100,
num_elite=10,
std=0.3,
ensemble_num=5)
policy_net.set_params(mean_params)
print(get_params(env))
online_osi(env,
osi,
policy_net,
num_init_traj=1,
max_horizon=15,
eval_episodes=10,
use_state=True,
print_timestep=10,
resample_MP=resample_MP,
online=0,
ensemble=5)
from model import make
import numpy as np
import sys
import os
# os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
Dp = 10
Hp = 175
Wp = 175
T = 35
model = make(Dp, Hp, Wp, T)
model.summary()
model.compile('sgd', ['mse', 'mse'])
#Dp = 10
#Hp = 400
#Wp = 352
#T = 35
# TODO: data preparation, i.e. grouping and sampling (this is done outside the network)
x = np.random.random((1, Dp, Hp, Wp, T, 7))
y1 = np.random.random((1, Hp // 2, Wp // 2, 2))
y2 = np.random.random((1, Hp // 2, Wp // 2, 14))
def gen():
yield x, [y1, y2]
import tensorflow as tf
import model
import shutil
# Kerasのモデルを読み込む
model = model.make(tflite=True)
# ニューラルの重みを読み込む
model.load_weights("weight.hdf5")
# TensorFlowのセッションを取得
sess = tf.keras.backend.get_session()
# SavedModelを出力
shutil.rmtree("saved_model/", True)
tf.saved_model.simple_save(sess,
"saved_model/",
inputs={'input': model.inputs[0]},
outputs={'output': model.outputs[0]})