-
Notifications
You must be signed in to change notification settings - Fork 2
/
core.py
232 lines (198 loc) · 8.47 KB
/
core.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
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
# -*- coding: utf-8 -*-
from keras.models import load_model
from keras import optimizers
from generator import DataGenerator, PredictDataGenerator,\
getTimePeriod,ncFileDir_2016,ncFileDir_2017,M,npyWRFFileDir, getHoursGridFromNC, \
getHoursGridFromNPY, num_frames, wrf_fea_dim, \
param_list, fea_dim, GuiTruthGridDir, num_frames_truth
import os
import numpy as np
import datetime
from keras import backend as K
import keras.backend.tensorflow_backend as KTF
import tensorflow as tf
from keras.callbacks import Callback
from keras.callbacks import ModelCheckpoint
import new_models
import scores
modelfileDir = 'models/'
def POD(y_true, y_pred):
ytrue = K.flatten(y_true)
ypred = K.sigmoid(K.flatten(y_pred))
ypred = K.round(ypred)
true_positives = K.sum(ytrue * ypred)
possible_positives = K.sum(ytrue)
recall = true_positives / (possible_positives + K.epsilon())
return recall
def FAR(y_true, y_pred):
ytrue = K.flatten(y_true)
ypred = K.sigmoid(K.flatten(y_pred))
ypred = K.round(ypred)
true_positives = K.sum(ytrue * ypred)
predicted_positives = K.sum(ypred)
precision = true_positives / (predicted_positives + K.epsilon())
return 1 - precision
def TS(y_true, y_pred):
ytrue = K.flatten(y_true)
ypred = K.sigmoid(K.flatten(y_pred))
ypred = K.round(ypred)
N1 = K.sum(ytrue * ypred)
N1pN2 = K.sum(ypred)
N1pN3 = K.sum(ytrue)
N2 = N1pN2 - N1
N3 = N1pN3 - N1
TS = N1 / (N1 + N2 + N3 + K.epsilon())
return TS
def weight_loss(y_true,y_pred):
pw = 25
ytrue = K.flatten(y_true)
ypred = K.flatten(y_pred)
return tf.reduce_mean(tf.nn.weighted_cross_entropy_with_logits(logits=ypred,targets=ytrue,pos_weight=pw))
def binary_acc(y_true,y_pred):
ypred = K.sigmoid(y_pred)
return K.mean(K.equal(y_true, K.round(ypred)), axis=-1)
class RecordMetricsAfterEpoch(Callback):
def on_epoch_end(self, epoch, logs={}):
filename = modelrecordname
with open('records/' + filename + '.txt','a') as f:
f.write('epoch %d:\r\n' % (epoch+1))
for key in ['loss', 'POD', 'FAR', 'TS', 'binary_acc', 'val_loss','val_POD', 'val_FAR', 'val_TS']:
f.write('%s: %f ' % (key, logs[key]))
def DoTrain(train_list, val_list):
# parameters
train_batchsize = 4
val_batchsize = 1
class_num = 2
epochs_num = 30
initial_epoch_num = 0
# when train a new model -----------------------------------------
#model = new_models.ADSNet_Plain()
#model = new_models.ADSNet_W()
#model = new_models.ADSNet_O()
model = new_models.ADSNet()
#model = new_models.StepDeep_model()
# print(model.summary())
dt_now = datetime.datetime.now().strftime('%Y%m%d%H%M')
print(dt_now)
adam = optimizers.adam(lr=0.0001)
model.compile(
loss=weight_loss,
optimizer=adam,
metrics=[POD,FAR,TS,binary_acc])
modelfilename = "%s-%s-{epoch:02d}.hdf5" % (dt_now, model.name)
global modelrecordname
modelrecordname = dt_now + '_' + model.name
checkpoint = ModelCheckpoint(modelfileDir + modelfilename, monitor='val_loss', verbose=1,
save_best_only=False, mode='min')
train_gen = DataGenerator(train_list, train_batchsize, class_num, generator_type='train')
val_gen = DataGenerator(val_list, val_batchsize, class_num, generator_type='val')
RMAE = RecordMetricsAfterEpoch()
model.fit_generator(train_gen,
validation_data=val_gen,
epochs=epochs_num,
initial_epoch=initial_epoch_num,
# use_multiprocessing=True,
workers=3,
max_queue_size=50,
callbacks = [RMAE,checkpoint]
)
def DoTest_step_seq(test_list, model, modelfilepath, testset_disp):
test_batchsize = 1
M = 1
test_gen = PredictDataGenerator(test_list, test_batchsize)
print('generating test data and predicting...')
ypred = model.predict_generator(test_gen, workers=3, verbose=1) # [len(test_list),num_frames,159*159,1]
ypred = 1.0 / (1.0 + np.exp(-ypred)) # if model doesn't include a sigmoid layer
## plot (the prediction for timesteps) ------------------------------------
with tf.device('/cpu:0'):
for id, ddt_item in enumerate(test_list):
ddt = datetime.datetime.strptime(ddt_item, '%Y%m%d%H%M')
utc = ddt + datetime.timedelta(hours=-8) # convert Beijing time into UTC time
ft = utc + datetime.timedelta(hours=(-6) * M)
nchour, delta_hour = getTimePeriod(ft)
delta_hour += M * 6
y_pred = ypred[id] # [num_frames,159*159,1]
for hour_plus in range(num_frames):
y_pred_i = y_pred[hour_plus]
dt = ddt + datetime.timedelta(hours=hour_plus)
dt_item = dt.strftime('%Y%m%d%H%M')
resDir = 'results/%s_set%s/' % (modelfilepath, testset_disp)
if not os.path.isdir(resDir):
os.makedirs(resDir)
with open(resDir + '%s_h%d' % (dt_item, hour_plus), 'w') as rfile:
for i in range(159*159):
rfile.write('%f\r\n' % y_pred_i[i]) # the probability value
# print(dt_item)
def Test_att_weights(test_list, model, name):
test_batchsize = 1
test_gen = PredictDataGenerator(test_list, test_batchsize)
alpha_lists = model.predict_generator(test_gen, workers=3, verbose=1)
alpha_lists = np.array(alpha_lists)
param_list = ['QICE_ave3_%d' % i for i in range(9)] + \
['QSNOW_ave3_%d' % i for i in range(9)] + \
['QGRAUP_ave3_%d' % i for i in range(9)] + ['W_max'] + ['RAINNC']
import csv
with open('att_weights_%s.csv' % name, 'w', newline='') as file:
csv_writer = csv.writer(file)
for i in range(alpha_lists.shape[0]):
csv_writer.writerow(['case %d:' % i,])
tmp = ['hour-%d' % j for j in range(1, 13)]
tmp.insert(0,'')
tmp.append('ave')
csv_writer.writerow(tmp)
for k in range(alpha_lists.shape[2]):
tmp = list(alpha_lists[i,:,k])
ave = np.average(alpha_lists[i,:,k])
tmp.insert(0,param_list[k])
tmp.append(ave)
csv_writer.writerow(tmp)
csv_writer.writerow(['total average'])
tmp = ['hour-%d' % j for j in range(1, 13)]
tmp.insert(0, '')
tmp.append('ave')
for k in range(alpha_lists.shape[2]): # alpha_list (sample_num, 12, fea_dim)
cases_ave = np.average(alpha_lists, axis=0) # (12,fea_dim)
tmp = list(cases_ave[:,k])
ave = np.average(cases_ave[:,k])
tmp.insert(0, param_list[k])
tmp.append(ave)
csv_writer.writerow(tmp)
return
if __name__ == "__main__":
mode = 'TRAIN'
# mode = 'TEST'
os.environ['CUDA_VISIBLE_DEVICES'] = '0'
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
config.allow_soft_placement = True
sess = tf.Session(config=config)
KTF.set_session(sess)
print('num_frames', num_frames)
TrainSetFilePath = 'train_lite_new_12h.txt'
ValSetFilePath = 'July.txt'
TestSetFilePath = '20170809_6n.txt'
testset_disp = '20170809_6n'
if mode == 'TRAIN':
train_list = []
with open(TrainSetFilePath, 'r') as file:
for line in file:
train_list.append(line.rstrip('\n'))
val_list = []
with open(ValSetFilePath, 'r') as file:
for line in file:
val_list.append(line.rstrip('\n'))
DoTrain(train_list, val_list)
elif mode == 'TEST':
test_list = []
with open(TestSetFilePath, 'r') as file:
for line in file:
test_list.append(line.rstrip('\n'))
for i in [14]:
modelfilepath = '201908272358-WRF-ADSNet-%s.hdf5' % str(i).zfill(2)
trained_model = load_model(modelfileDir + modelfilepath,
{'weight_loss3': weight_loss, 'POD': POD, 'TS': TS,
'FAR': FAR, 'binary_acc': binary_acc, 'num_frames': num_frames})
DoTest_step_seq(test_list, trained_model, modelfilepath, testset_disp)
resultfolderpath = modelfilepath + '_set%s' % testset_disp
scores.eva(resultfolderpath, 0.5)
sess.close()