def get_stock_price(self, content):
name_to_ticker = {
"amazon": "amzn",
"google": "goog",
"chevron": "cvx",
"yahoo": "yhoo",
"walmart": "wal",
}
s_content = content.split()
ticker = s_content[-1]
if ticker in ("ticker", "stock") and len(ticker) >= 2:
ticker = s_content[-2]
if ticker in name_to_ticker:
ticker = name_to_ticker[ticker]
try:
price = web.get_last_iex(ticker).to_dict()[0]["price"]
self.respond(
"{} is at {}, this information has 15 minute delay".format(
ticker, price))
except Exception as exc:
print("Could not get price for ticker: {}".format(ticker))
def download_stock(stock):
""" try to query the iex for a stock, if failed note with print """
""" set the download window """
now_time = datetime.datetime.now()
past_five_pm=now_time.replace(hour=18, minute=0, second=0, microsecond=0)
start_time = datetime.datetime(now_time.year - 5, now_time.month , now_time.day)
weekno=datetime.datetime.today().weekday()
try:
print(stock)
stock_df=''
if(IEX==True):
stock_df = web.DataReader(stock,'iex', start_time, now_time)
stock_df['Name'] = stock
else:
stock_df = web.DataReader(stock,'robinhood', start_time, now_time)
output_name = os.path.join('.','stocks',stock+'_data.csv')
stock_df.to_csv(output_name)
if(now_time>=past_five_pm and weekno<5):
#get the last value padding ohlcv
if(IEX==True):
current=web.get_last_iex(stock)[0][0]
temphigh=stock_df['close'][len(stock_df)-1]
templow=stock_df['close'][len(stock_df)-1]
previous_close=stock_df['close'][len(stock_df)-1]
else:
current = float(web.get_quotes_robinhood(stock)[stock][7])
temphigh=float(stock_df['close_price'][len(stock_df)-1])
templow=float(stock_df['close_price'][len(stock_df)-1])
previous_close=float(stock_df['close_price'][len(stock_df)-1])
tempvol=int(stock_df['volume'][len(stock_df)-1])
current_date=datetime.datetime.today().strftime('%Y-%m-%d')
if(current>temphigh):
temphigh=current
if(templow>current):
templow=current
#df.append({ 'date':current_date, 'open':previous_close, 'high':temphigh, 'low':templow, 'close':current, 'volume':tempvol, 'Name':stock })
with open(os.path.join('.','stocks',stock+'_data.csv'), 'a') as f:
if(IEX==True):
f.write( '%s,%0.2f,%0.2f,%0.2f,%0.2f,%d,%s\n'%(current_date, previous_close, temphigh, templow, current, tempvol, stock))
else:
f.write( '%s,%s,%0.6f,%0.6f,False,%0.6f,%0.6f,reg,%d\n'%(stock, current_date, current, temphigh, templow, previous_close, tempvol))
except Exception as e:
bad_names.append(stock)
print('bad: %s %s' % (e,stock))
def getQuoteGoogle():
tickers = ['AAPL', 'MSFT', '^GSPC']
start_date = '2010-01-01'
end_date = '2016-12-31'
panel_data = web.DataReader('INPX', 'google', start_date, end_date)
web.get_last_iex()
def test_live_prices(self):
dftickers = get_iex_symbols()
tickers = dftickers[:5].symbol.values
df = get_last_iex(tickers[:5])
assert df["price"].mean() > 0
def test_false_ticker(self):
df = get_last_iex("INVALID TICKER")
assert df.shape[0] == 0
def test_false_ticker(self):
df = get_last_iex("INVALID TICKER")
tm.assert_frame_equal(df, DataFrame())
def do_main(argv):
stock_dataset = ''
if (len(argv) >= 3 and argv[2] == '-d'):
stock = argv[1]
now_time = datetime.datetime.now()
past_five_pm = now_time.replace(hour=18,
minute=0,
second=0,
microsecond=0)
weekno = datetime.datetime.today().weekday()
stock_dataset = web.DataReader(
argv[1], 'iex',
datetime.datetime(datetime.datetime.now().year - 5,
datetime.datetime.now().month,
datetime.datetime.now().day),
datetime.datetime.now() + datetime.timedelta(days=1))
stock_dataset['Name'] = stock
stock_dataset.to_csv('./stocks/' + argv[1] + '_data.csv')
if (now_time >= past_five_pm and weekno < 5):
#get the last value padding ohlcv
current = web.get_last_iex(stock)[0][0]
temphigh = stock_dataset['close'][len(stock_dataset) - 1]
templow = stock_dataset['close'][len(stock_dataset) - 1]
tempvol = int(stock_dataset['volume'][len(stock_dataset) - 1])
previous_close = stock_dataset['close'][len(stock_dataset) - 1]
current_date = datetime.datetime.today().strftime('%Y-%m-%d')
if (current > temphigh):
temphigh = current
if (templow > current):
templow = current
with open(os.path.join('.', 'stocks', stock + '_data.csv'),
'a') as f:
f.write('%s,%0.2f,%0.2f,%0.2f,%0.2f,%d,%s\n' %
(current_date, previous_close, temphigh, templow,
current, tempvol, stock))
stock_dataset = pd.read_csv('./stocks/' + argv[1] + '_data.csv')
else:
stock_dataset = pd.read_csv('./stocks/' + argv[1] + '_data.csv')
dataset = stock_dataset['close']
dataset2 = stock_dataset['low']
#x_scaler = MinMaxScaler(feature_range=(0, 1))
#y_scaler = MinMaxScaler(feature_range=(0, 1))
window_size = 10
X = []
y = []
for i in range(window_size, len(dataset)):
X.append(dataset[i - window_size:i])
for i in range(window_size, len(dataset2)):
y.append([dataset2[i]])
X.append(dataset[len(dataset) - window_size:])
X = np.asarray(X)
y = np.asarray(y)
#X=x_scaler.fit_transform(X)
#y_train=y_scaler.fit_transform(y)
y_train = y
X_train = X[:-1, ]
X_test = X[-1:, ]
# NOTE: to use keras's RNN LSTM module our input must be reshaped to [samples, window size, stepsize]
X_train = np.asarray(
np.reshape(X_train, (X_train.shape[0], window_size, 1)))
X_test = np.asarray(np.reshape(X_test, (X_test.shape[0], window_size, 1)))
# start with fixed random seed
np.random.seed(0)
# Build an RNN to perform regression on our time series input/output data
model = Sequential()
model.add(LSTM(128, input_shape=(window_size, 1), return_sequences=True))
model.add(LSTM(64, return_sequences=True))
model.add(LSTM(32))
model.add(Dense(16))
model.add(Dense(1))
optimizer = keras.optimizers.RMSprop(lr=0.001,
rho=0.9,
epsilon=1e-08,
decay=0.0)
model.summary()
# compile the model
model.compile(loss='mean_squared_error', optimizer=optimizer)
model.fit(X_train, y_train, epochs=500, batch_size=256, verbose=1)
# generate predictions for training
test_predict = model.predict(X_test)
#prediction=y_scaler.inverse_transform(test_predict)[0][0]
prediction = test_predict[0][0]
dataset = np.asarray(dataset2)
history = dataset[-1]
print("Prediction %s: %f vs Last Value: %f" %
(argv[1], prediction, history))
if (prediction > history):
return True
else:
return False