dl4jModelimport.java

  File completeModelFile = new ClassPathResource("complete.h5").getFile();
  String completeModelFileName = completeModelFile.getAbsolutePath();
  MultiLayerNetwork network = KerasModelImport.importKerasSequentialModelAndWeights(completeModelFileName);

  NativeImageLoader loader = new NativeImageLoader(height, width);
  File imageFile = new ClassPathResource("1.jpg").getFile();

  INDArray image = loader.asMatrix(imageFile);
  image.divi(255); // <== how about that? :)
  INDArray output = network.output(image);

keras_1.2.2_theano_model.py

import numpy as np
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation, Flatten
from keras.layers import Convolution2D, MaxPooling2D
from keras import backend as K
from scipy.misc import imread
from keras.models import load_model, save_model
np.random.seed(1000)  # for reproducibility

nb_classes = 10
img_rows, img_cols = 24, 15
nb_filters = 32
pool_size = (2, 2)
kernel_size = (3, 3)

img = imread("./1.jpg")
X_test = img.reshape(1, img_rows, img_cols)

if K.image_dim_ordering() == 'th':
    X_test = X_test.reshape(X_test.shape[0], 1, img_rows, img_cols)
    input_shape = (1, img_rows, img_cols)
else:
    X_test = X_test.reshape(X_test.shape[0], img_rows, img_cols, 1)
    input_shape = (img_rows, img_cols, 1)

X_test = X_test.astype('float32')
X_test /= 255

model = Sequential()

model.add(Convolution2D(nb_filters, kernel_size[0], kernel_size[1],
                        border_mode='valid',
                        input_shape=input_shape))
model.add(Activation('relu'))
model.add(Convolution2D(nb_filters, kernel_size[0], kernel_size[1]))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=pool_size))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))

model.compile(loss='categorical_crossentropy',
              optimizer='adadelta',
              metrics=['accuracy'])

model_json = model.to_json()
with open("keras_model.json", "w") as f:
    f.write(model_json)
model.save_weights("keras_weights.h5")
print("Saved model to disk")

save_model(model, "./complete.h5")
print(pred)

output_comparison

dl4j:  [0.12, 0.11, 0.09, 0.10, 0.08, 0.10, 0.09, 0.10, 0.10, 0.10]
keras: [0.11, 0.11, 0.09, 0.11, 0.06, 0.10, 0.09, 0.11, 0.10  0.09]