In [9]:

    

from keras.models import Sequential
from keras.layers import Dense
from keras.models import model_from_json
import numpy as np
import os
import cv2
%matplotlib inline


    

In [5]:

    

testLoc = '../data/test1/'
data = [os.path.join(testLoc,f) for f in os.listdir(testLoc)]


    

In [7]:

    

def imageToFeatures(x,size=(32,32)):
    return cv2.resize(x,size).flatten()

testX = [imageToFeatures(cv2.imread(x)) for x in data]


    

In [10]:

    

testX = np.array(testX)


    

In [11]:

    

testX = testX/255.0


    

In [12]:

    

json_file = open('model.json', 'r')
loaded_model_json = json_file.read()
json_file.close()
loaded_model = model_from_json(loaded_model_json)
# load weights into new model
loaded_model.load_weights("model.h5")
print("Loaded model from disk")


    

    




Loaded model from disk

In [20]:

    

output = loaded_model.predict_proba(batch_size=32,x=testX)


    

    




12480/12500 [============================>.] - ETA: 0s

In [23]:

    

output.shape


    

    
Out[23]:





(12500, 2)

In [26]:

    

import csv
with open('keras_simple_nn.csv','w') as csvFile:
    writer = csv.writer(csvFile)
    writer.writerow(['id','label'])
    for i,value in enumerate(output[:,1]):
        writer.writerow([i+1,value])


    

In [27]:

    

#from matplotlib import pyplot as plt
#plt.imshow(plt.imread(data[2]))


    

In [ ]: