How to run Keras.model() for prediction inside a tensorflow session?

Question

I am currently having an issue, while executing my model predict of keras inside a tensorflow session.

with tf.Session(graph=graph) as sess:
    sess.run(tf.global_variables_initializer())
    ## want to know how to add model.predict() inside this condition
    predictions = model.predict(#my_model)
    #predictions output is same not appending

or any alternative method will be helpful.

Any help would be appreciated.

Answer 1

from keras import backend as K

with tf.Graph().as_default():

    with tf.Session() as sess:

        K.set_session(sess)
        model = load_model(model_path)
        preds = model.predict(in_data)

Answer 2

from keras.models import load_model
with tf.Session(graph=K.get_session().graph) as session:
    session.run(tf.global_variables_initializer())
    model = load_model('model.h5')
    predictions = model.predict(input)

Above code works for me. I am using keras mobilenet inside tensorflow.