Deep Learning Example (Python Implementation) requires the file (prima-indians-diabetes.csv) available as a separate gist.

Sugar_Diabetes_Predictions.py

# my first neural network with keras
# requires that TensorFlow or Theano be installed
from numpy import loadtxt
from keras.models import Sequential
from keras.layers import Dense
from keras.models import model_from_json
import numpy
import os

# fix random seed for reproducibility
numpy.random.seed(7)

# load the data into a Dataset
#There are eigth input variables:
#
# input Variable (x):
#
# 1. Number of times pregnant.
# 2. Plasma glucose concentration, 2 hours in an Oral glucose tolerance test.
# 3. Diastolic blood pressure (mm Hg).
# 4. Triceps skin fold thickness (mm).
# 5. 2-Hour Serum Insulin (mu U/ml).
# 6. Body mass index (weigth in kg/(height in m)^2).
# 7. Diabetes pedigree function.
# 8. Age (Years).
#
# Output Variable (y):
#
# 1. Class variable (0 or 1)

dataset = loadtxt('prima-indians-diabetes.csv', delimiter=',')

save_dir = os.path.join(os.getcwd(), 'saved_models')
model_name = 'keras_Sugar_Predictions_trained_model.h5'
model_json_name = "Keras_Prima.json"

# we will train the system 30 times to determine its accuracy.
for i in range(30):	
	# split into input (x) and output (y) variables
	x = dataset[:,0:8]
	y = dataset[:,8]
	
	# define the keras model
	model = Sequential()
	model.add(Dense(12, input_dim=8, activation='relu'))
	model.add(Dense(8, activation='relu'))
	model.add(Dense(1, activation='sigmoid'))
	
	# compile the keras model
	model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
	
	# fit the keras model on the dataset
	# we will use 200 epochs and batch size of 10 for improved Accuracy of (upto %80.2.)
	model.fit(x, y, epochs=200, batch_size=10, verbose=0) #set verbose to 1 if you want the training progress displayed.
	# evaluate the keras model
	_, accuracy = model.evaluate(x, y)
	print("Accuracy: %{0}".format(accuracy*100))
	
	# we will save if our desired accuracy is reached.
	if ((accuracy*100) > 79.50):
		# Save model and weights
		# serialize model to JSON
		model_json = model.to_json()
		with open(os.path.join(save_dir, model_json_name), "w") as json_file:
			json_file.write(model_json)
		# serialize weights to HDF5
		model_path = os.path.join(save_dir,model_name)
		model.save_weights(model_path)
		print("Saved trained model to disk")
		break # stop the loop.

# make class predictions with the model
#
# split into input (x) and output (y) variables
x = dataset[:,0:8]
y = dataset[:,8]

# Load from disk
model_json_path = os.path.join(save_dir,model_json_name)
model_path = os.path.join(save_dir,model_name)

if (os.path.exists(model_path) and os.path.exists(model_json_path)):
	
	#load json and create model
	json_file = open(model_json_path, "r")
	loaded_model_json = json_file.read()
	json_file.close()
	model2 = model_from_json(loaded_model_json)
	
	# load weights into new model
	model2.load_weights(model_path)
	print("Loaded model from disk")
	model2.compile(loss="binary_crossentropy", optimizer="rmsprop", metrics=["accuracy"])
	predictions = model2.predict_classes(x)
	# summarize the first five cases
	for i in range(5):
		print('%s => %d (expected %d)' % (x[i].tolist(), predictions[i], y[i]))