notebook.community

Edit and run 





In [92]:



    


import numpy
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import LSTM
from keras.layers.embeddings import Embedding
from keras.preprocessing import sequence
import pandas as pd
from sklearn import cross_validation

# load data
top_words = 4214 # insert max index of items + 1
max_length = 10 # length of sequential data
df = pd.read_csv("c:/dev/dl/purchacedata_base.csv", header=0)

df.head()



    


















    
Out[92]:










  
    

      
      Y
      CustomerID
      P1
      P2
      P3
      P4
      P5
      P6
      P7
      P8
      ...
      P91
      P92
      P93
      P94
      P95
      P96
      P97
      P98
      P99
      P100
    

  
  
    

      0
      0
      12346
      604
      604.0
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      ...
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
    

    

      1
      0
      12348
      120
      238.0
      60.0
      60.0
      109.0
      109.0
      181.0
      553.0
      ...
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
    

    

      2
      0
      12349
      603
      2259.0
      1333.0
      649.0
      1072.0
      488.0
      344.0
      140.0
      ...
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
    

    

      3
      0
      12350
      217
      1117.0
      1073.0
      1847.0
      384.0
      2139.0
      102.0
      148.0
      ...
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
      NaN
    

    

      4
      0
      12352
      1155
      630.0
      779.0
      1325.0
      402.0
      348.0
      524.0
      1839.0
      ...
      749.0
      537.0
      1066.0
      1196.0
      11.0
      NaN
      NaN
      NaN
      NaN
      NaN
    

  



5 rows × 102 columns



















In [93]:



    


# extract columns and drop rows having NaN
df = df[df.columns[0:(max_length + 2)]].dropna().astype(int)

train_X, test_X, train_Y, test_Y = cross_validation.train_test_split(df[df.columns[2:(max_length + 2)]], df["Y"])

train_X = train_X.values
train_Y = train_Y.values
test_X = test_X.values
test_Y = test_Y.values

print(train_X.shape)
print(test_X.shape)
print(test_Y)



    


















    






(2835, 10)
(946, 10)
[0 0 0 0 0 1 1 1 1 0 0 0 0 1 0 0 1 1 0 1 0 0 0 0 1 0 0 0 1 0 0 1 0 0 1 0 0
 0 0 0 0 0 1 1 1 0 0 1 0 1 1 1 1 1 1 1 0 1 1 0 1 0 0 0 1 1 1 0 1 0 0 1 0 0
 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 1 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 1 0 0
 1 1 1 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 1 0 0 0 0 0 1 0 0 0 1 1 0 0
 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 1 0 0 0 0 0 1 0 0 0 1 0 0 0
 0 1 0 0 0 0 1 0 0 1 0 0 0 0 0 1 0 0 0 0 1 0 1 0 0 0 1 0 0 0 1 0 1 0 0 0 0
 1 0 1 0 1 0 0 0 0 1 0 0 0 0 1 0 1 1 1 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 0 1
 1 0 1 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 1 0 0 0 0 0 1 1 0 1 1 0 1 1 0 0
 0 1 0 0 1 0 1 0 0 0 0 0 0 0 0 0 1 1 0 0 0 1 0 0 1 0 0 0 1 1 1 1 0 0 1 1 1
 0 0 0 0 0 0 1 0 0 0 0 0 1 0 1 0 1 0 1 0 0 0 0 1 0 0 0 1 0 0 1 0 1 0 1 0 1
 1 0 1 0 0 0 1 0 0 0 1 0 0 0 1 1 0 1 0 1 1 1 1 0 0 0 0 1 0 0 0 0 1 0 0 0 1
 0 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 0 1 0 1 0 1 0
 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 0 0 1 0 0 1 0 0
 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 1 0 1 0 1 0 0 1 0 0 1 0 1 0 0 0 0 1
 0 1 0 0 0 1 1 1 0 0 0 0 0 0 1 1 1 0 1 1 0 0 0 0 0 0 1 0 0 1 0 0 0 1 0 0 1
 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 1 0 1 0 1 0 0 0
 0 1 1 0 0 0 0 0 0 1 1 0 1 1 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0
 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0 1 1 1 0 1 1 0 1 0 1 0 0 0 1 0 0 0 0 0 0 0
 0 0 0 0 1 0 0 0 1 0 0 0 0 1 0 0 0 1 0 1 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 1 0
 0 1 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 0
 0 0 0 0 0 0 0 0 1 0 0 1 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 1 0 0 0 1 1 0 0
 1 1 0 0 0 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 1 1 0 0 0 1 0 1 0 0 0 0 1 0 0 1 1
 0 1 0 1 1 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 0
 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 1 1 0 0 0 0 1
 1 1 0 1 0 1 1 0 0 0 0 1 0 0 0 0 1 0 1 0 1]

















In [107]:



    


# create the model
embedding_vector_length = 5
model = Sequential()
model.add(Embedding(top_words, embedding_vector_length, input_length=max_length))
model.add(LSTM(3))
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
print(model.summary())
model.fit(train_X, train_Y, validation_data=(test_X, test_Y), epochs=3, batch_size=64)



    


















    






_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
embedding_29 (Embedding)     (None, 10, 5)             21070     
_________________________________________________________________
lstm_29 (LSTM)               (None, 3)                 108       
_________________________________________________________________
dense_29 (Dense)             (None, 1)                 4         
=================================================================
Total params: 21,182
Trainable params: 21,182
Non-trainable params: 0
_________________________________________________________________
None
Train on 2835 samples, validate on 946 samples
Epoch 1/3
2835/2835 [==============================] - 1s - loss: 0.6792 - acc: 0.6942 - val_loss: 0.6594 - val_acc: 0.7241
Epoch 2/3
2835/2835 [==============================] - 0s - loss: 0.6411 - acc: 0.7012 - val_loss: 0.6106 - val_acc: 0.7241
Epoch 3/3
2835/2835 [==============================] - 0s - loss: 0.6002 - acc: 0.7012 - val_loss: 0.5859 - val_acc: 0.7241










    
Out[107]:








<keras.callbacks.History at 0x5e178668>

















In [108]:



    


# Final evaluation of the model
scores = model.evaluate(test_X, test_Y, verbose=0)
print("Accuracy: %.2f%%" % (scores[1]*100))



    


















    






Accuracy: 72.41%

















In [ ]:

Content source: tizuo/keras

Similar notebooks:

notebook.community | gallery | about