forked from holmescao/iGrow
-
Notifications
You must be signed in to change notification settings - Fork 0
/
evaluate_simulator.py
192 lines (165 loc) · 6.96 KB
/
evaluate_simulator.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
import pickle
import pandas as pd
import numpy as np
import argparse
import os
import warnings
import torch
import gym
from sklearn.metrics import r2_score
from TenSim.utils.data_reader import TomatoDataset
from TenSim.simulator import PredictModel
from utils.common import mkdir
warnings.filterwarnings("ignore")
os.environ['NLS_LANG'] = 'AMERICAN_AMERICA.AL32UTF8'
gym.logger.set_level(40)
torch.set_num_threads(1)
def env(version, base_tmp_folder):
direcrory = base_tmp_folder+'/models/'
model_dir = direcrory + version
model_path = model_dir + '/model/'
scaler_dir = model_dir + '/scaler/'
ten_env = PredictModel(model1_dir=model_path+'simulator_greenhouse.pkl',
model2_dir=model_path+'simulator_crop_front.pkl',
model3_dir=model_path+'simulator_crop_back.pkl',
scaler1_x=scaler_dir+'greenhouse_x_scaler.pkl',
scaler1_y=scaler_dir+'greenhouse_y_scaler.pkl',
scaler2_x=scaler_dir+'crop_front_x_scaler.pkl',
scaler2_y=scaler_dir+'crop_front_y_scaler.pkl',
scaler3_x=scaler_dir+'crop_back_x_scaler.pkl',
scaler3_y=scaler_dir+'crop_back_y_scaler.pkl',
linreg_dir=model_path+'/PARsensor_regression_paramsters.pkl',
weather_dir=model_path+'/weather.npy')
return ten_env
def Table1(args):
print("=============Table1===============")
tmp_folder = os.path.join(args.base_tmp_folder,
'models/%s' % args.model_version)
wur_tomato_reader = TomatoDataset(args.traj_test_files, tmp_folder)
train_data = wur_tomato_reader.read_data(args.traj_test_files)
full_train_x, _ = wur_tomato_reader.data_process(train_data)
simulator = env(args.model_version, args.base_tmp_folder)
# PAR model
PAR_model_path = os.path.join(
tmp_folder, 'model/PARsensor_regression_paramsters.pkl')
linreg = pickle.load(open(PAR_model_path, 'rb'))
columns = ['PAR', 'AirT', 'AirRh', 'AirCO2',
'LAI', 'PlantLoad', ' NetGrouwth', 'FW']
save_dir = args.base_tmp_folder + '/table1/%s/' % args.model_version
mkdir(save_dir)
save_path = save_dir+'R2_of_per_cache.csv'
if os.path.exists(save_path):
os.remove(save_path)
PAR_R2 = []
AirT_R2 = []
AirRH_R2 = []
Airppm_R2 = []
LAI_R2 = []
PlantLoad_R2 = []
NetGrowth_R2 = []
FW_R2 = []
score = []
for idx in range(len(full_train_x)):
PAR_list, real_PAR_list = [], []
AirT_list, real_AirT_list = [], []
AirRH_list, real_AirRH_list = [], []
Airppm_list, real_Airppm_list = [], []
LAI_list, real_LAI_list = [], []
PlantLoad_list, real_PlantLoad_list = [], []
NetGrowth_list, real_NetGrowth_list = [], []
FW_list, real_FW_list = [], []
input = full_train_x[idx]
done = False
simulator.reset()
for i in range(args.DAY_IN_LIFE_CYCLE):
control = input[i * 24: (i + 1) * 24, 6: 10]
control = control.T.reshape(1, -1)[0]
obs, _, done, _ = simulator.step(control)
if done:
break
for h in range(24):
par_x = input[i*24 + h: i*24 + h+1, [0, 8]].reshape(1, -1)
PARsensor = linreg.predict(par_x)
PARsensor = float(PARsensor) if PARsensor > 50.0 else 0.0
PAR_list.append(PARsensor)
real_PAR_list.extend(input[i*24: (i+1)*24, 13])
AirT_list.extend(obs[: 24])
real_AirT_list.extend(
input[i * 24:(i + 1) * 24, 10].reshape(1, -1).tolist()[0])
AirRH_list.extend(obs[24:48])
real_AirRH_list.extend(
input[i * 24:(i + 1) * 24, 11].reshape(1, -1).tolist()[0])
Airppm_list.extend(obs[48:72])
real_Airppm_list.extend(
input[i * 24:(i + 1) * 24, 12].reshape(1, -1).tolist()[0])
LAI_list.append(obs[72])
real_LAI_list.append(input[i * 24 + 23, 14])
PlantLoad_list.append(obs[73])
real_PlantLoad_list.append(input[i * 24 + 23, 15])
NetGrowth_list.append(obs[74])
real_NetGrowth_list.append(input[i * 24 + 23, 16])
FW_list.append(obs[75])
real_FW_list.append(input[i * 24 + 23, 17])
# calculate R^2
r2_PAR = r2_score(real_PAR_list, PAR_list)
r2_AirT = r2_score(real_AirT_list, AirT_list)
r2_AirRH = r2_score(real_AirRH_list, AirRH_list)
r2_Airppm = r2_score(real_Airppm_list, Airppm_list)
r2_LAI = r2_score(real_LAI_list, LAI_list)
r2_PlantLoad = r2_score(real_PlantLoad_list, PlantLoad_list)
r2_NetGrowth = r2_score(real_NetGrowth_list, NetGrowth_list)
r2_FW = r2_score(real_FW_list, FW_list)
goodness = [r2_PAR, r2_AirT, r2_AirRH, r2_Airppm,
r2_LAI, r2_PlantLoad, r2_NetGrowth,
r2_FW]
mean_r2 = np.mean(goodness)
goodness.append(mean_r2)
print("%d cache score: %.2f" % (idx, mean_r2))
# # save
df = pd.DataFrame([goodness], columns=columns+['score'])
if os.path.exists(save_path):
ori_df = pd.read_csv(save_path)
df = ori_df.append(df)
df.to_csv(save_path, float_format='%.3f', index=False)
# net1
PAR_R2.append(r2_PAR)
AirT_R2.append(r2_AirT)
AirRH_R2.append(r2_AirRH)
Airppm_R2.append(r2_Airppm)
# net2
LAI_R2.append(r2_LAI)
PlantLoad_R2.append(r2_PlantLoad)
NetGrowth_R2.append(r2_NetGrowth)
# net3
FW_R2.append(r2_FW)
score.append(mean_r2)
# mean
mean_PAR = np.mean(PAR_R2)
mean_AirT = np.mean(AirT_R2)
mean_AirRH = np.mean(AirRH_R2)
mean_Airppm = np.mean(Airppm_R2)
mean_LAI = np.mean(LAI_R2)
mean_PlantLoad = np.mean(PlantLoad_R2)
mean_NetGrowth = np.mean(NetGrowth_R2)
mean_FW = np.mean(FW_R2)
mean_score = np.mean(score)
goodness_of_simulator = [mean_PAR, mean_AirT, mean_AirRH, mean_Airppm,
mean_LAI, mean_PlantLoad, mean_NetGrowth,
mean_FW, mean_score]
# save
Table1_df = pd.DataFrame([goodness_of_simulator],
columns=columns+['score'])
Table1_df.to_csv(save_dir+'R2_of_simulator.csv',
float_format='%.3f', index=False)
print("mean R2:")
print(Table1_df.mean(axis=0))
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument("--base_tmp_folder", default="./result", type=str)
parser.add_argument("--model_version", default="baseline", type=str)
parser.add_argument("--traj_test_files",
default="./input/test-sim.txt", type=str)
parser.add_argument("--DAY_IN_LIFE_CYCLE",
default=160, type=int)
args = parser.parse_args()
Table1(args)