# Main project file for stock price prediction
from fetchdata import fetch
from partitiondata import partition_split
import numpy as np
import matplotlib.pyplot as plt
from scipy.interpolate import spline
from sklearn.linear_model.ridge import Ridge
from sklearn.linear_model import RidgeCV
from sklearn.feature_selection import RFE
from sklearn.feature_selection import RFECV
# Load CSV file
alldata = fetch('../data/stock_returns_base150.csv')
alldata.pop(0); # Remove the header
print(len(alldata));
# Partition data into separate objects for training, testing and forecasting. Also generate the features matrix X and output y
lag = 3;
print("Lag is ", lag);
testpercent = 25;
n_modeldata = 50;
X,Xtest,y,ytest,Xfcast,features = partition_split(alldata,n_modeldata,testpercent,lag);
print(' X is ', X.shape , '\t Xtest is ', Xtest.shape, '\t Xfcast is ', Xfcast.shape,
'\n y is ',y.shape , '\t \t ytest is ', ytest.shape, '\n features is ', features.shape);
n_features = features.shape[0];
# Features Ranking
# Univariate Feature Scoring with Cross validation
def plotscores(scores,rf,n = n_features):
def dataReceived(self, data):
data = data.decode("utf-8")
if data == "refresh":
self.write(fetchdata.fetch())
