def add_numbers():
a = request.args.get('a', 0, type=str)
hashvalue = ''.join(
random.choice(string.ascii_uppercase + string.digits)
for _ in range(6))
#r = test_1()
#print r
data = get_data(a, hashvalue)
data = [tuple(x) for x in data.to_records(index=True)]
cur, conn = build_connection()
run_sql(cur, conn, data)
print jsonify(result=hashvalue)
return jsonify(result=hashvalue)
# -*- coding: utf-8 -*- """ @author: Asma Baccouche """ from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer from Get_Data import get_data from sklearn.cluster import KMeans, SpectralClustering from nltk.cluster import KMeansClusterer, util from gensim.models import Word2Vec from sklearn import metrics import numpy as np from Data_helper import sent_vectorizer data, deps, Deps_Count = get_data() sentences = [sentence for sentence in data['TITLE']] s = [sentence.split() for sentence in data['TITLE']] vectorizer1 = CountVectorizer() vectorizer2 = TfidfVectorizer() X_TF = vectorizer1.fit_transform(sentences) X_TFIDF = vectorizer2.fit_transform(sentences) kmeans1 = KMeans(n_clusters=8).fit(X_TF) labels1 = kmeans1.labels_ kmeans2 = KMeans(n_clusters=8).fit(X_TFIDF) labels2 = kmeans2.labels_ SpectralClustering1 = SpectralClustering(n_clusters=8,
import numpy as np from Get_Data import get_data tick=get_data(1) tick['return']=abs(tick.close-tick.open)/tick.open vol=np.random.normal(tick[-90:]['return'].mean(),tick[-90:]['return'].std(),5) ud=[1,-1,1,1,-1] price=25.2 vollist=(abs(vol)*ud+1) print (vollist) print(price*np.cumprod(vollist))
import numpy as np
import pandas as pd
import random
from sklearn import preprocessing
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
from Get_Data import get_data
##load data and calculate label
all_data = get_data(1)
all_data['close_1'] = all_data.close.shift(-1)
all_data['open_1'] = all_data.open.shift(-1)
all_data['stdev'] = (all_data.close - all_data.open) / all_data.open
all_data = all_data[:-1]
labeltest = []
def cal_label(shift_no=1):
for i in range(len(all_data)):
##two class
if all_data.close.shift(-shift_no)[i] / all_data.open.shift(
-shift_no)[i] > (all_data.stdev.quantile(.55) + 1.000):
labeltest.append([1, 0, 0])
elif all_data.close.shift(-shift_no)[i] / all_data.open.shift(
-shift_no)[i] < (all_data.stdev.quantile(.45) + 1.000):
labeltest.append([0, 1, 0])
else:
labeltest.append([0, 0, 1])
all_data['label'] = pd.Series(labeltest, index=all_data.index)
from Get_Data import get_data
from sklearn import preprocessing
import pandas as pd
import numpy as np
from keras.models import Sequential
from keras.layers.core import Dense,Dropout,Activation
from keras.layers.recurrent import LSTM
import keras
raw_data= get_data(1)
raw_data=raw_data[['open','high','low','volume','close']]
raw_data.dropna(how='any',inplace=True)
##data scaler
def normalize(df):
new_data=df.copy()
min_max_scaler=preprocessing.MinMaxScaler()
new_data['open']=min_max_scaler.fit_transform(raw_data.open.values.reshape(-1,1))
new_data['high']=min_max_scaler.fit_transform(raw_data.high.values.reshape(-1,1))
new_data['low']=min_max_scaler.fit_transform(raw_data.low.values.reshape(-1,1))
new_data['volume']=min_max_scaler.fit_transform(raw_data.volume.values.reshape(-1,1))
new_data['close']=min_max_scaler.fit_transform(raw_data.close.values.reshape(-1,1))
return new_data
##classfy multiple set
new_data=normalize(raw_data)
No_feature=len(new_data.columns)
new_matrix=new_data.as_matrix()