Keras for implement convolution neural network

As Matias says in the comments, this is pretty straightforward... Keras updated their API yesterday to 2.0 version. Obviously you have downloaded that version and the demo still uses the "old" API. They have created warnings so that the "old" API would still work in the version 2.0, but saying that it will change so please use 2.0 API from now on.

The way to adapt your code to API 2.0 is to change the "init" parameter to "kernel_initializer" for all of the Dense() layers as well as the "nb_epoch" to "epochs" in the fit() function.

from keras.models import Sequential
from keras.layers import Dense
import numpy
# fix random seed for reproducibility
seed = 7
numpy.random.seed(seed)
# load pima indians dataset
dataset = numpy.loadtxt("pima-indians-diabetes.csv", delimiter=",")
# split into input (X) and output (Y) variables
X = dataset[:,0:8]
Y = dataset[:,8]
# create model
model = Sequential()
model.add(Dense(12, input_dim=8, kernel_initializer ='uniform', activation='relu'))
model.add(Dense(8, kernel_initializer ='uniform', activation='relu'))
model.add(Dense(1, kernel_initializer ='uniform', activation='sigmoid'))
# Compile model
model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
# Fit the model
model.fit(X, Y, epochs=10, batch_size=10)
# evaluate the model
scores = model.evaluate(X, Y)
print("%s: %.2f%%" % (model.metrics_names[1], scores[1]*100))

This shouldn't throw any warnings, it's the keras 2.0 version of the code.