-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy pathpointnet_large_separate_valid.py
170 lines (128 loc) · 6.65 KB
/
pointnet_large_separate_valid.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
import sys
import os
import argparse
import pandas as pd
import datetime as dt
import numpy as np
from keras.models import Sequential
from keras.layers import Dense, Conv1D, Flatten, Reshape, Dropout, Activation, BatchNormalization
from keras.callbacks import EarlyStopping, ModelCheckpoint, CSVLogger
from keras.utils import multi_gpu_model
from keras.optimizers import Adam
import ideal_meta
import ideal_loader
from data_provider import S2PDataProvider
def load_train_data():
# load train data
y_train, x_train, t_train, _ \
= ideal_loader.load_appliances([appliance], train_homes, sample_rate=sample_rate)
y_valid, x_valid, t_valid, _ \
= ideal_loader.load_appliances([appliance], valid_homes, sample_rate=sample_rate)
return x_train, y_train, t_train, x_valid, y_valid, t_valid
def load_test_data(home):
# load data
y_test, x_test, t_test, house_appliances \
= ideal_loader.load_appliances([appliance], [home], sample_rate=sample_rate)
return x_test, y_test, t_test
def normalise_inputs(inputs_batch):
return (inputs_batch - inputs_batch.mean(axis=1).reshape((-1,1))) / ideal_meta.mains_std
def normalise_targets(targets_batch):
return (targets_batch - mean_on_power) / std_on_power
def predictions_transform(predictions):
return predictions * std_on_power + mean_on_power
def create_model(input_window_length):
model = Sequential()
model.add(Reshape([input_window_length, 1], input_shape=[input_window_length]))
model.add(Conv1D(30, 10, padding='same', activation='relu'))
model.add(Dropout(0.5))
model.add(Conv1D(30, 8, padding='same', activation='relu'))
model.add(Dropout(0.5))
model.add(Conv1D(40, 6, padding='same', activation='relu'))
model.add(Dropout(0.5))
model.add(Conv1D(50, 5, padding='same', activation='relu'))
model.add(Dropout(0.5))
model.add(Conv1D(50, 5, padding='same', activation='relu'))
model.add(Dropout(0.5))
model.add(Flatten())
model.add(Dense(1024, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(1, activation=None))
return model
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='Train pointnet model on IDEAL')
parser.add_argument('--appliance', default='kettle')
parser.add_argument('--predict_only', action='store_true')
args = parser.parse_args()
sample_rate = 8
gpus = 1
appliance = args.appliance
#refit_appliance = ideal_loader.appliance_ideal2refit[appliance]
train_homes = np.genfromtxt('train_homes.csv').astype(int)
valid_homes = np.genfromtxt('valid_homes.csv').astype(int)
test_homes = np.genfromtxt('test_homes.csv').astype(int)
batch_size = 1024 * gpus
input_window_length = 599#refit_meta.pointnet_window_length[refit_appliance]
IDEAL_DATA_DIR = os.environ['IDEAL_DATA_DIR']
model_name = 'pointnet_large_separate_valid'
appliance_model_name = model_name + '_{0}'.format(appliance)
model_path = IDEAL_DATA_DIR + '/models/' + appliance_model_name + '.h5'
stats_path = IDEAL_DATA_DIR + '/stats/' + appliance_model_name +'.csv'
appliance_stats = pd.read_csv('appliance_stats.csv')
appliance_stats = appliance_stats[appliance_stats.appliancetype==appliance]
mean_on_power = float(appliance_stats.mean_on_power)
std_on_power = float(appliance_stats.std_on_power)
# create keras model
model = create_model(input_window_length)
if gpus > 1:
model = multi_gpu_model(model, gpus=gpus)
if args.predict_only:
model.load_weights(model_path)
else:
# use maximum 8 weeks of data from each home
history_length = int(dt.timedelta(weeks=8).total_seconds() / sample_rate)
x_train, y_train, t_train, x_valid, y_valid, t_valid = load_train_data()
train_provider = S2PDataProvider(x_train, y_train, t_train, batch_size=batch_size,
input_window_size=input_window_length,
shuffle_order=True, inputs_transform=normalise_inputs,
targets_transform=normalise_targets)
valid_provider = S2PDataProvider(x_valid, y_valid, t_valid, batch_size=batch_size,
input_window_size=input_window_length,
shuffle_order=True, inputs_transform=normalise_inputs,
targets_transform=normalise_targets)
print('Training samples: {0}'.format(train_provider.indices.shape))
print('Validation examples: {0}'.format(valid_provider.indices.shape))
model.compile(loss='mean_squared_error', optimizer=Adam(lr=0.001))
model.summary()
callbacks = [ModelCheckpoint(model_path, monitor='val_loss', save_best_only=True,
save_weights_only=True),
EarlyStopping(monitor='val_loss', patience=10),
CSVLogger(stats_path)]
model.fit_generator(train_provider.generator(), train_provider.num_batches, epochs=100,
verbose=1, validation_data=valid_provider.generator(),
validation_steps=valid_provider.num_batches, callbacks=callbacks)
# make predictions
predictions_store = pd.HDFStore(IDEAL_DATA_DIR + '/predictions/{0}.h5'.format(model_name))
for home in test_homes:
x_test, y_test, t_test = load_test_data(home)
if len(x_test) > 0:
results_name = '/home_{0}/{1}'.format(home, appliance)
print(results_name)
test_provider = S2PDataProvider(x_test, y_test, t_test, batch_size=batch_size,
input_window_size=input_window_length,
shuffle_order=False, inputs_transform=normalise_inputs)
predictions = model.predict_generator(test_provider.generator(),
test_provider.num_batches, verbose=1)
predictions = predictions.flatten()
predictions = predictions_transform(predictions)
# clip negative values
predictions[predictions < 0] = 0
t_test = t_test[0][test_provider.indices]
x_test = x_test[0][test_provider.indices]
y_test = y_test[0][test_provider.indices]
# put results into dataframe and store
results = pd.DataFrame(
index=pd.to_datetime(t_test, unit='s'),
data={'mains':x_test,
'predictions':predictions,
'ground_truth':y_test})
predictions_store[results_name] = results