tfserver 0.3.3

Tensor Flow Model Server

Introduce

tfserver is an example for serving Tensorflow model with Skitai App Engine.

It can be accessed by gRPC and JSON RESTful API.

This project is inspired by issue #176.

From version 0.3, it is now TensorFlow 2+ compatible.

Table of Contents

• Introduce

• Saving Tensorflow Model

• Running Server

  • Adding Custom APIs

• Request Examples

  • gRPC Client

  • RESTful API

• Performance Note Comparing with Proto Buffer and JSON

  • Test Environment

  • Results

• Release History

Saving Tensorflow Model

# MUST import top of the code for disabling eager execution
from tfserver import saved_model
import tensorflow as tf
import numpy as np

x = tf.keras.layers.Input (3)
h = tf.keras.layers.Dense (10, activation='relu') (x)
y = tf.keras.layers.Dense (2, activation='softmax') (h)

model = tf.keras.Model (x, y)
model.compile (
    optimizer=tf.keras.optimizers.Adam(0.001),
    loss = 'categorical_crossentropy',
    metrics = ['accuracy']
)

model.summary()
train_data = np.array ([
    (0.1, 0.2, 0.6),
    (0.3, 0.6, 0.7),
    (0.2, 0.9, 0.3),
    (0.3, 0.9, 0.1),
])

labels = np.array ([
    (1.0, 0),
    (0, 1.0),
    (1.0, 0),
    (0, 1.0),
])

model.fit(train_data, labels, epochs=3, batch_size=32)
model.evaluate(train_data, labels)
model.predict(train_data)
inputs, outputs = saved_model.save ('exported', {'x': model.inputs[0]}, {'y': model.outputs[0]}, 'predict')

print ("* Saved Model")
print ("  - Inputs")
for k, v in inputs.items (): print ("    . {}: {}".format (k, v.name))
print ("  - Outputs")
for k, v in outputs.items (): print ("    . {}: {}".format (k, v.name))

Running Server

You just setup model path and tensorflow configuration, then you can have gRPC and JSON API services.

Example of api.py

import tfserver
import skitai

pref = skitai.pref ()
pref.max_client_body_size = 100 * 1024 * 1024 # 100 MB

# we want to serve 2 models:
# alias and (model_dir, optional session config)
tfserver.add_model ("model1", "exported/model1/200", gpu_usage = 0.1)
tfserver.add_model ("model1", "exported/model2/100", 0.2)

# If you want to activate gRPC, should mount on '/'
skitai.mount ("/", tfserver, pref = pref)
skitai.run (port = 5000)

And run,

python3 api.py

Adding Custom APIs

You can create your own APIs.

For example,

# services/apis.py

import tfserver

def predict (alias, signature_name, **inputs):
    result = tfserver.run (alias, signature_name, **inputs)
    pred = np.argmax (result ["y"][0])
    return dict (
        confidence = float (result ["y"][0][pred]),
        code = tfserver.tfsess [alias].labels [0].item (pred)
    )

def __mount__ (app):
    import os
    from dnn import tf
    from .helpers.unspsc import datautil

    def initialize_models (app):
        for alias, (model_dir, gpu_usage) in tfserver.added_models.items ():
            if model == "f22":
                datautil.load_features (os.path.join (model_path, 'features.pkl'))
    initialize_models (app)

    @app.route ("/", methods = ["GET"])
    def models (was):
        return was.API (models = list (tfserver.tfsess.keys ()))

    @app.route ("/unspsc", methods = ["POST"])
    def unspsc (was, text, signature_name = "predict"):
        x, seq_length = datautil.encode (text)
        result = predict ("unspsc", signature_name, x = [x], seq_length = [seq_length])
        return was.API (result = result)

Then mount these services and run.

# serve.py

from services import apis

with skitai.preference () as pref:
    pref.mount ("/apis", apis)
          skitai.mount ("/", tfserver, pref = pref)
      skitai.run (port = 5000, name = "tfapi")

Request Examples

gRPC Client

Using grpcio library,

from tfserver import cli
from tensorflow.python.framework import tensor_util
import numpy as np

stub = cli.Server ("http://localhost:5000")
problem = np.array ([1.0, 2.0])

resp = stub.predict (
  'model1', #alias for model
  'predict', #signature_def_name
  x = tensor_util.make_tensor_proto(problem.astype('float32'), shape=problem.shape)
)
# then get 'y'
resp.y
>> np.ndarray ([-1.5, 1.6])

Using aquests for async request,

import aquests
from tfserver import cli
from tensorflow.python.framework import tensor_util
import numpy as np

def print_result (resp):
  cli.Response (resp.data).y
  >> np.ndarray ([-1.5, 1.6])

stub = aquests.grpc ("http://localhost:5000/tensorflow.serving.PredictionService", callback = print_result)
problem = np.array ([1.0, 2.0])

request = cli.build_request (
  'model1',
  'predict',
  x = problem
)
stub.Predict (request, 10.0)

aquests.fetchall ()

RESTful API

Using requests,

import requests

problem = np.array ([1.0, 2.0])
api = requests.session ()
resp = api.post (
  "http://localhost:5000/predict",
  json.dumps ({"x": problem.astype ("float32").tolist()}),
  headers = {"Content-Type": "application/json"}
)
data = json.loads (resp.text)
data ["y"]
>> [-1.5, 1.6]

Another,

from aquests.lib import siesta

problem = np.array ([1.0, 2.0])
api = siesta.API ("http://localhost:5000")
resp = api.predict.post ({"x": problem.astype ("float32").tolist()})
resp.data.y
>> [-1.5, 1.6]

Performance Note Comparing with Proto Buffer and JSON

Test Environment

• Input:

  • dtype: Float 32

  • shape: Various, From (50, 1025) To (300, 1025), Prox. Average (100, 1025)

• Output:

  • dtype: Float 32

  • shape: (60,)

• Request Threads: 16

• Requests Per Thread: 100

• Total Requests: 1,600

Results

Average of 3 runs,

• gRPC with Proto Buffer:

  • Use grpcio

  • 11.58 seconds

• RESTful API with JSON

  • Use requests

  • 216.66 seconds

Proto Buffer is 20 times faster than JSON…

Release History

• 0.3 (2018. 6. 28): reactivate project and compatible with TF2+

• 0.2 (2020. 6. 26): integrated with dnn 0.3

• 0.1b8 (2018. 4. 13): fix grpc trailers, skitai upgrade is required

• 0.1b6 (2018. 3. 19): found works only grpcio 1.4.0

• 0.1b3 (2018. 2. 4): add @app.umounted decorator for clearing resource

• 0.1b2: remove self.tfsess.run (tf.global_variables_initializer())

• 0.1b1 (2018. 1. 28): Beta release

• 0.1a (2018. 1. 4): Alpha release