In [ ]:
import os
import sys
# This is a placeholder for a Google-internal import.
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)
# Base path for exporting model
# Set it to /home/jovyan/tf-serving-base/models to easily create a container and deploy
export_path_base = "/home/jovyan/tf-serving-base/models"
# Model Version
model_version = 1
In [ ]:
from tensorflow.python.client import device_lib
local_device_protos = device_lib.list_local_devices()
[print(x.name) for x in local_device_protos if x.device_type == 'GPU']
In [ ]:
# Training iterations
training_iteration = 1000
print('Training model...')
sess = tf.InteractiveSession()
serialized_tf_example = tf.placeholder(tf.string, name='tf_example')
feature_configs = {'x': tf.FixedLenFeature(shape=[784], dtype=tf.float32),}
tf_example = tf.parse_example(serialized_tf_example, feature_configs)
x = tf.identity(tf_example['x'], name='x') # use tf.identity() to assign name
y_ = tf.placeholder('float', shape=[None, 10])
w = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))
sess.run(tf.global_variables_initializer())
y = tf.nn.softmax(tf.matmul(x, w) + b, name='y')
cross_entropy = -tf.reduce_sum(y_ * tf.log(y))
train_step = tf.train.GradientDescentOptimizer(0.01).minimize(cross_entropy)
values, indices = tf.nn.top_k(y, 10)
table = tf.contrib.lookup.index_to_string_table_from_tensor(
tf.constant([str(i) for i in range(10)]))
prediction_classes = table.lookup(tf.to_int64(indices))
for _ in range(training_iteration):
batch = mnist.train.next_batch(50)
train_step.run(feed_dict={x: batch[0], y_: batch[1]})
correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, 'float'))
print('training accuracy %g' % sess.run(
accuracy, feed_dict={x: mnist.test.images,
y_: mnist.test.labels}))
print('Done training!')
In [ ]:
export_path = os.path.join(
tf.compat.as_bytes(export_path_base),
tf.compat.as_bytes(str(model_version)))
print('Exporting trained model to', export_path)
builder = tf.saved_model.builder.SavedModelBuilder(export_path)
# Build the signature_def_map.
classification_inputs = tf.saved_model.utils.build_tensor_info(serialized_tf_example)
classification_outputs_classes = tf.saved_model.utils.build_tensor_info(prediction_classes)
classification_outputs_scores = tf.saved_model.utils.build_tensor_info(values)
classification_signature = (
tf.saved_model.signature_def_utils.build_signature_def(
inputs={
tf.saved_model.signature_constants.CLASSIFY_INPUTS:
classification_inputs
},
outputs={
tf.saved_model.signature_constants.CLASSIFY_OUTPUT_CLASSES:
classification_outputs_classes,
tf.saved_model.signature_constants.CLASSIFY_OUTPUT_SCORES:
classification_outputs_scores
},
method_name=tf.saved_model.signature_constants.CLASSIFY_METHOD_NAME))
tensor_info_x = tf.saved_model.utils.build_tensor_info(x)
tensor_info_y = tf.saved_model.utils.build_tensor_info(y)
prediction_signature = (
tf.saved_model.signature_def_utils.build_signature_def(
inputs={'images': tensor_info_x},
outputs={'scores': tensor_info_y},
method_name=tf.saved_model.signature_constants.PREDICT_METHOD_NAME))
legacy_init_op = tf.group(tf.tables_initializer(), name='legacy_init_op')
builder.add_meta_graph_and_variables(
sess, [tf.saved_model.tag_constants.SERVING],
signature_def_map={
'predict_images':
prediction_signature,
tf.saved_model.signature_constants.DEFAULT_SERVING_SIGNATURE_DEF_KEY:
classification_signature,
},
legacy_init_op=legacy_init_op)
builder.save()
print('Done exporting!')
In [ ]:
from os import path
import yaml
import time
from kubernetes import client, config
config.load_incluster_config()
with open("/home/jovyan/tf-serving-base/deployment.yaml") as f:
dep = yaml.load(f)
# Specify the newly built image.
dep['spec']['template']['spec']['containers'][0]['image'] = "gcr.io/vishnuk-cloud/tf-model-server:latest"
k8s_beta = client.ExtensionsV1beta1Api()
try:
resp = k8s_beta.delete_namespaced_deployment(body={}, namespace="jupyterhub", name=dep['metadata']['name'])
time.sleep(10)
except:
pass
resp = k8s_beta.create_namespaced_deployment(body=dep, namespace="jupyterhub")
print("Deployment created. status='%s'" % str(resp.status))
with open("/home/jovyan/tf-serving-base/service.yaml") as f:
dep = yaml.load(f)
v1 = client.CoreV1Api()
v1.delete_namespaced_service(namespace="jupyterhub", name=dep['metadata']['name'])
resp = v1.create_namespaced_service(body=dep, namespace="jupyterhub")
print("Service created. status='%s'" % str(resp.status))
In [ ]: