foo.py

# coding: utf-8

# In[ ]:





# In[1]:


get_ipython().magic('matplotlib inline')
import numpy as np
import matplotlib.pyplot as plt

dataset = np.loadtxt('normalized_ibm_prices.csv')

get_ipython().magic('matplotlib inline')
import numpy as np
import matplotlib.pyplot as plt

plt.plot(dataset)
plt.xlabel('time period')
plt.ylabel('normalized series value')


# In[4]:


def window_transform_series(series,window_size,idx = 3):
    # containers for input/output pairs
    X = []
    y = []

    upper_range = len(series)-window_size
    for i in range(0,upper_range):               
        window_index = i+window_size
        X_temp = [series[j] for j in range(i,window_index) ]
        X.append(X_temp)        
        y.append( series[i+window_size:i+window_size+idx])                        
        
    # reshape each 
    X = np.asarray(X)
    dir((np.shape(X)[0:2]))    
    X.shape = (np.shape(X)[0:2])
    y = np.asarray(y)
    y.reshape(len(y),1)
    return X,y 


window_size = 7
X,y = window_transform_series(dataset,7)
print (X.shape)
print (y.shape)
print(X[0])
print(y[0])


# In[6]:


# split our dataset into training / testing sets
train_test_split = int(np.ceil(2*len(y)/float(3)))   # set the split point

# partition the training set
X_train = X[:train_test_split,:]
y_train = y[:train_test_split]

# keep the last chunk for testing
X_test = X[train_test_split:,:]
y_test = y[train_test_split:]

# 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)))


# In[7]:


from keras.models import Sequential
from keras.layers import Dense
from keras.layers import LSTM
from keras.callbacks import EarlyStopping
import keras

# given - fix random seed - so we can all reproduce the same results on our default time series
np.random.seed(0)


# TODO: build an RNN to perform regression on our time series input/output data

model = Sequential()
model.add(LSTM(5, input_shape = (window_size,1)))
model.add(Dense(1, activation=None))

print(len(X_train))
print(len(y_train))

# build model using keras documentation recommended optimizer initialization
optimizer = keras.optimizers.RMSprop(lr=0.001, rho=0.9, epsilon=1e-08, decay=0.0)

# compile the model
model.compile(loss='mean_squared_error', optimizer=optimizer)

model.fit(X_train, y_train, epochs=100, verbose=2)