stocks.py

import numpy as np
from numpy import newaxis
import matplotlib.pyplot as plt
import pandas as pd
from pandas import datetime
import math, time
import itertools
from sklearn import preprocessing
import datetime
from operator import itemgetter
from sklearn.metrics import mean_squared_error
from math import sqrt
from keras.models import Sequential
from keras.layers.core import Dense, Dropout, Activation
from keras.layers.recurrent import LSTM
import time

def is_number(num):
	for a in str(num):
		if not a.isdigit() and a != '.':
			return 0
	return float(num)
	

start = time.time()
stock_name = 'FB'

url="http://www.google.com/finance/historical?q="+stock_name+"&startdate=Jul+12%2C+2013&enddate=Jul+31%2C+2017&num=30&ei=rCtlWZGSFN3KsQHwrqWQCw&output=csv"
col_names = ['Date','Open','High','Low','Close','Volume']
stocks = pd.read_csv(url, header=0, names=col_names) 
df = pd.DataFrame(stocks)
df.drop(df.columns[[0,3,5]], axis=1, inplace=True) 
df.tail()

today = datetime.date.today()
file_name = stock_name+'_stock_%s.csv' % today
#df.to_csv(file_name)

df['High'] = [is_number(a) for a in df['High']]
df['Open'] = [is_number(a) for a in df['Open']]
df['Close'] = [is_number(a) for a in df['Close']]

print(df['Open'][0])

df['High'] = [a / 1000 for a in df['High']]
for a in df['Open']:
	print(a)
	print(a/1000)
df['Open'] = [a / 1000 for a in df['Open']]
df['Close'] = [a / 1000 for a in df['Close']]
df.head(5)

stock = df[::-1]
seq_len = 5
amount_of_features = len(stock.columns)
data = stock.as_matrix() #pd.DataFrame(stock)
print(data)
sequence_length = seq_len + 1
result = []
for index in range(len(data) - sequence_length):
    result.append(data[index: index + sequence_length])

result = np.array(result)
row = round(result.shape[0]) - 95
if result.shape[0] < 1000:
	row = round(result.shape[0]) - 7
train = result[:int(row), :]
x_train = train[:, :-1]
y_train = train[:, -1][:,-1]
x_test = result[int(row):, :-1]
y_test = result[int(row):, -1][:,-1]

x_train = np.reshape(x_train, (x_train.shape[0], x_train.shape[1], amount_of_features))
x_test = np.reshape(x_test, (x_test.shape[0], x_test.shape[1], amount_of_features))  

print("X_train", x_train.shape)
print("y_train", y_train.shape)
print("X_test", x_test.shape)
print("y_test", y_test.shape)

d = 0.2
model = Sequential()
model.add(LSTM(128, input_shape=(x_train.shape[1], x_train.shape[2]), return_sequences=True))
model.add(Dropout(d))
model.add(LSTM(64, input_shape=(x_train.shape[1], x_train.shape[2]), return_sequences=True))
model.add(Dropout(d))
model.add(LSTM(32, input_shape=(x_train.shape[1], x_train.shape[2]), return_sequences=False))
model.add(Dropout(d))
model.add(Dense(16,init='uniform',activation='relu'))        
model.add(Dense(1,init='uniform',activation='relu'))
model.compile(loss='mse',optimizer='adam')

model.fit(
    x_train,
    y_train,
    batch_size=512,
    nb_epoch=300,
    validation_split=0.1,
    verbose=1)

#trainScore = 
print(model.evaluate(x_train, y_train, verbose=0))
#print('Train Score: %.2f MSE (%.2f RMSE)' % (trainScore[0], math.sqrt(trainScore[0])))

#testScore = model.evaluate(x_test, y_test, verbose=0)
#print('Test Score: %.2f MSE (%.2f RMSE)' % (testScore[0], math.sqrt(testScore[0])))

p = model.predict(x_test)

import matplotlib.pyplot as plt2
end = time.time()

print(end - start)
plt2.plot(p[5:],color='red', label='predicted')
plt2.plot(y_test[:-5],color='blue', label='actual')
plt2.legend(loc='upper left')
plt.title(stock_name)
plt2.show()