YAO/YAO-optimizer
1
0
Fork 0
mirror of https://github.com/newnius/YAO-optimizer.git synced 2025-12-13 00:16:44 +00:00
Code Issues Releases Wiki Activity

update

Browse Source
This commit is contained in:
Newnius
2020-05-01 16:35:56 +08:00
parent a394657a54
commit c68a1aaa5a
3 changed files with 140 additions and 144 deletions
Download Patch File Download Diff File Expand all files Collapse all files

109
serve.py
View File

@@ -11,11 +11,13 @@ import os
from sklearn.model_selection import train_test_split
from model_tensorflow import train, predict
import csv
from collections import deque
from StringIO import StringIO
class Config:
feature_columns = list(range(0, 6))
label_columns = [3, 4, 5]
feature_columns = list([2, 5])
label_columns = [5]
feature_and_label_columns = feature_columns + label_columns
label_in_feature_columns = (lambda x, y: [x.index(i) for i in y])(feature_columns, label_columns)
@@ -34,7 +36,8 @@ class Config:
add_train = False
shuffle_train_data = True
train_data_rate = 0.95
# train_data_rate = 0.95 #comment yqy
train_data_rate = 1 # add yqy
valid_data_rate = 0.15
batch_size = 64
@@ -50,10 +53,10 @@ class Config:
batch_size = 1
continue_flag = "continue_"
train_data_path = "./data.csv"
test_data_path = "./test.csv"
train_data_path = "./data/data.csv"
model_save_path = "./checkpoint/"
figure_save_path = "./figure/"
do_figure_save = False
if not os.path.exists(model_save_path):
os.mkdir(model_save_path)
@@ -74,17 +77,14 @@ class Data:
self.train_num = int(self.data_num * self.config.train_data_rate)
self.mean = np.mean(self.data, axis=0)
self.std = np.std(self.data, axis=0)
self.norm_data = (self.data - self.mean) / self.std
self.start_num_in_test = 0
def read_data(self):
init_data = pd.read_csv(
self.config.train_data_path,
usecols=self.config.feature_and_label_columns
)
init_data = pd.read_csv(self.config.train_data_path,
usecols=self.config.feature_and_label_columns)
return init_data.values, init_data.columns.tolist()
def get_train_and_valid_data(self):
@@ -112,22 +112,6 @@ class Data:
return train_x, valid_x, train_y, valid_y
def get_test_data(self, return_label_data=False):
init_data = pd.read_csv(
self.config.test_data_path,
usecols=self.config.feature_and_label_columns
)
norm_data = (init_data - self.mean) / self.std
feature_data = norm_data[:]
time_step_size = feature_data.shape[0] // self.config.time_step
test_x = [feature_data[0 + i * self.config.time_step: 0 + (
i + 1) * self.config.time_step]
for i in range(time_step_size)]
if return_label_data:
label_data = norm_data[0:, self.config.label_in_feature_columns]
return np.array(test_x), label_data
return np.array(test_x)
feature_data = self.norm_data[self.train_num:]
self.start_num_in_test = feature_data.shape[0] % self.config.time_step
time_step_size = feature_data.shape[0] // self.config.time_step
@@ -140,31 +124,46 @@ class Data:
return np.array(test_x), label_data
return np.array(test_x)
# add yqy
def get_test_data_yqy(self, test_data_yqy=None):
if test_data_yqy is None:
test_data_yqy = []
# test_data_yqy=test_data_yqy[1:21]
feature_data = (test_data_yqy - self.mean) / self.std
test_x = [feature_data]
return np.array(test_x)
def draw(config, origin_data, predict_norm_data):
label_norm_data = origin_data.norm_data[origin_data.train_num + origin_data.start_num_in_test:,
config.label_in_feature_columns]
# add end
def draw_yqy(config, origin_data, predict_norm_data, mean_yqy, std_yqy): # 这里origin_data等同于test_data_values_yqy
label_norm_data = (origin_data - mean_yqy) / std_yqy
assert label_norm_data.shape[0] == predict_norm_data.shape[
0], "The element number in origin and predicted data is different"
label_name = [origin_data.data_column_name[i] for i in config.label_in_feature_columns]
label_column_num = len(config.label_columns)
# label_norm_data=label_norm_data[:,1]
label_name = 'high'
label_column_num = 1
loss = np.mean((label_norm_data[config.predict_day:] - predict_norm_data[:-config.predict_day]) ** 2, axis=0)
loss = \
np.mean((label_norm_data[config.predict_day:][:, 1] - predict_norm_data[:-config.predict_day][0:]) ** 2,
axis=0)[1]
print("The mean squared error of stock {} is ".format(label_name), loss)
label_X = range(origin_data.data_num - origin_data.train_num - origin_data.start_num_in_test)
predict_X = [x + config.predict_day for x in label_X]
# label_X = range(origin_data.data_num - origin_data.train_num - origin_data.start_num_in_test)
# predict_X = [x + config.predict_day for x in label_X]
label_data = label_norm_data * origin_data.std[config.label_in_feature_columns] + \
origin_data.mean[config.label_in_feature_columns]
label_data = label_norm_data[:, 1] * std_yqy[1] + mean_yqy[1]
predict_data = predict_norm_data * origin_data.std[config.label_in_feature_columns] + \
origin_data.mean[config.label_in_feature_columns]
predict_data = predict_norm_data * std_yqy[1] + mean_yqy[1]
print(label_data)
print(predict_data)
print(label_data[-1])
print(predict_data[-1])
PORT_NUMBER = 8080
lock = Lock()
@@ -199,15 +198,35 @@ class MyHandler(BaseHTTPRequestHandler):
elif req.path == "/predict":
try:
job = query.get('job')[0]
gpu_model = query.get('gpu_model')[0]
time = query.get('time')[0]
data = {
'job': query.get('job')[0],
'model': query.get('model')[0],
'time': query.get('time')[0],
'utilGPU': query.get('utilGPU')[0],
'utilCPU': query.get('utilCPU')[0],
'pre': 0,
'main': 0,
'post': 0
}
with open(config.train_data_path, 'r') as f:
q = deque(f, config.time_step - 1)
df = pd.read_csv(StringIO(''.join(q)), usecols=config.feature_and_label_columns)
df.append(data, ignore_index=True)
df.to_csv('./data/test_data.csv')
np.random.seed(config.random_seed)
data_gainer = Data(config)
test_X, test_Y = data_gainer.get_test_data(return_label_data=True)
print(test_X, test_Y)
test_data_yqy = pd.read_csv("./data/test_data.csv", usecols=list(range(0, 8)))
test_data_values = test_data_yqy.values[:]
test_X = data_gainer.get_test_data_yqy(test_data_values)
pred_result = predict(config, test_X)
print(pred_result)
draw(config, data_gainer, pred_result)
mean = Data(config).mean
std = Data(config).std
draw_yqy(config, test_data_values, pred_result, mean, std)
msg = {'code': 1, 'error': "container not exist"}
except Exception as e:
msg = {'code': 2, 'error': str(e)}