Keras com Python

In [23]:

    

import pandas as pd
import numpy as np
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
import keras
from keras.models import Sequential
from keras.layers import Activation, Dense, Dropout, Flatten
from keras.optimizers import Adam
import matplotlib.pyplot as plt
%matplotlib inline


    

    




/home/55018335734/.conda/envs/datascience/lib/python3.6/site-packages/matplotlib/font_manager.py:280: UserWarning: Matplotlib is building the font cache using fc-list. This may take a moment.
  'Matplotlib is building the font cache using fc-list. '

In [2]:

    

df = pd.read_csv('evasao.csv')
df.head()


    

    
Out[2]:







  

    

      

      
periodo
      
bolsa
      
repetiu
      
ematraso
      
disciplinas
      
faltas
      
desempenho
      
abandonou
    
  
  

    

      
0
      
2
      
0.25
      
8
      
1
      
4
      
0
      
0.000000
      
1
    
    

      
1
      
2
      
0.15
      
3
      
1
      
3
      
6
      
5.333333
      
0
    
    

      
2
      
4
      
0.10
      
0
      
1
      
1
      
0
      
8.000000
      
0
    
    

      
3
      
4
      
0.20
      
8
      
1
      
1
      
0
      
4.000000
      
1
    
    

      
4
      
1
      
0.20
      
3
      
1
      
1
      
1
      
8.000000
      
0
    
  

In [3]:

    

df.describe()


    

    
Out[3]:







  

    

      

      
periodo
      
bolsa
      
repetiu
      
ematraso
      
disciplinas
      
faltas
      
desempenho
      
abandonou
    
  
  

    

      
count
      
300.000000
      
300.000000
      
300.000000
      
300.000000
      
300.000000
      
300.000000
      
300.000000
      
300.000000
    
    

      
mean
      
5.460000
      
0.123333
      
2.776667
      
0.476667
      
2.293333
      
2.213333
      
2.623111
      
0.410000
    
    

      
std
      
2.937772
      
0.086490
      
2.530111
      
0.500290
      
1.648133
      
2.734853
      
2.583423
      
0.492655
    
    

      
min
      
1.000000
      
0.000000
      
0.000000
      
0.000000
      
0.000000
      
0.000000
      
0.000000
      
0.000000
    
    

      
25%
      
3.000000
      
0.050000
      
0.000000
      
0.000000
      
1.000000
      
0.000000
      
0.400000
      
0.000000
    
    

      
50%
      
5.000000
      
0.100000
      
2.000000
      
0.000000
      
2.000000
      
1.000000
      
2.000000
      
0.000000
    
    

      
75%
      
8.000000
      
0.200000
      
5.000000
      
1.000000
      
4.000000
      
4.000000
      
4.000000
      
1.000000
    
    

      
max
      
10.000000
      
0.250000
      
8.000000
      
1.000000
      
5.000000
      
10.000000
      
10.000000
      
1.000000
    
  

In [4]:

    

features = df[['periodo','bolsa','repetiu','ematraso','disciplinas','faltas']]
labels = df[['abandonou']]


    

In [5]:

    

X_train, X_test, y_train, y_test = train_test_split(features, labels, test_size=0.33, random_state=42)


    

In [8]:

    

model = Sequential()


    

In [9]:

    

model.add(Dense(128, input_dim=6, kernel_initializer='normal', activation='relu'))
model.add(Dense(512, activation='relu'))
model.add(Dense(512, activation='relu'))
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='binary_crossentropy', optimizer='rmsprop', metrics=['accuracy'])


    

In [14]:

    

df_train_x = X_train.values
df_train_y = y_train.values
historico = model.fit(df_train_x, df_train_y)


    

    




Epoch 1/10
201/201 [==============================] - 0s 369us/step - loss: 0.5524 - acc: 0.7065
Epoch 2/10
201/201 [==============================] - 0s 241us/step - loss: 0.5460 - acc: 0.6816
Epoch 3/10
201/201 [==============================] - 0s 228us/step - loss: 0.5280 - acc: 0.7015
Epoch 4/10
201/201 [==============================] - 0s 195us/step - loss: 0.5278 - acc: 0.6766
Epoch 5/10
201/201 [==============================] - 0s 207us/step - loss: 0.5146 - acc: 0.7065
Epoch 6/10
201/201 [==============================] - 0s 229us/step - loss: 0.5156 - acc: 0.7065
Epoch 7/10
201/201 [==============================] - 0s 202us/step - loss: 0.5153 - acc: 0.7015
Epoch 8/10
201/201 [==============================] - 0s 219us/step - loss: 0.4922 - acc: 0.7065
Epoch 9/10
201/201 [==============================] - 0s 221us/step - loss: 0.5050 - acc: 0.7214
Epoch 10/10
201/201 [==============================] - 0s 235us/step - loss: 0.4835 - acc: 0.7612

In [15]:

    

model.summary()


    

    




_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_9 (Dense)              (None, 128)               896       
_________________________________________________________________
dense_10 (Dense)             (None, 512)               66048     
_________________________________________________________________
dense_11 (Dense)             (None, 512)               262656    
_________________________________________________________________
dense_12 (Dense)             (None, 1)                 513       
=================================================================
Total params: 330,113
Trainable params: 330,113
Non-trainable params: 0
_________________________________________________________________

In [26]:

    

print(historico.history.keys())
# summarize history for accuracy
plt.plot(historico.history['acc'])
#plt.plot(historico.history['val_acc'])
plt.title('model accuracy')
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper left')
plt.show()
# summarize history for loss
plt.plot(historico.history['loss'])
#plt.plot(historico.history['val_loss'])
plt.title('model loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['train', 'test'], loc='upper left')
plt.show()


    

    




dict_keys(['loss', 'acc'])






    













    

In [36]:

    

df_test_x = X_test.values
df_test_y = y_test.values
print(model.metrics_names)
model.evaluate(df_test_x,df_test_y)


    

    




['loss', 'acc']
99/99 [==============================] - 0s 127us/step






    
Out[36]:





[0.6022236907120907, 0.64646464706671358]