国产xxxx99真实实拍_久久不雅视频_高清韩国a级特黄毛片_嗯老师别我我受不了了小说

資訊專欄INFORMATION COLUMN

使用 TensorFlow Serving 和 Docker 快速部署機(jī)器學(xué)習(xí)服務(wù)

endless_road / 2664人閱讀

摘要：從實(shí)驗(yàn)到生產(chǎn)，簡單快速部署機(jī)器學(xué)習(xí)模型一直是一個挑戰(zhàn)。總結(jié)上面我們快速實(shí)踐了使用和部署機(jī)器學(xué)習(xí)服務(wù)的過程，可以看到，提供了非常方便和高效的模型管理，配合，可以快速搭建起機(jī)器學(xué)習(xí)服務(wù)。

從實(shí)驗(yàn)到生產(chǎn)，簡單快速部署機(jī)器學(xué)習(xí)模型一直是一個挑戰(zhàn)。這個過程要做的就是將訓(xùn)練好的模型對外提供預(yù)測服務(wù)。在生產(chǎn)中，這個過程需要可重現(xiàn)，隔離和安全。這里，我們使用基于Docker的TensorFlow Serving來簡單地完成這個過程。TensorFlow 從1.8版本開始支持Docker部署，包括CPU和GPU，非常方便。

獲得訓(xùn)練好的模型

獲取模型的第一步當(dāng)然是訓(xùn)練一個模型，但是這不是本篇的重點(diǎn)，所以我們使用一個已經(jīng)訓(xùn)練好的模型，比如ResNet。TensorFlow Serving 使用SavedModel這種格式來保存其模型，SavedModel是一種獨(dú)立于語言的，可恢復(fù)，密集的序列化格式，支持使用更高級別的系統(tǒng)和工具來生成，使用和轉(zhuǎn)換TensorFlow模型。這里我們直接下載一個預(yù)訓(xùn)練好的模型：

$ mkdir /tmp/resnet
$ curl -s https://storage.googleapis.com/download.tensorflow.org/models/official/20181001_resnet/savedmodels/resnet_v2_fp32_savedmodel_NHWC_jpg.tar.gz | tar --strip-components=2 -C /tmp/resnet -xvz

如果是使用其他框架比如Keras生成的模型，則需要將模型轉(zhuǎn)換為SavedModel格式，比如：

from keras.models import Sequential
from keras import backend as K
import tensorflow as tf

model = Sequential()
# 中間省略模型構(gòu)建

# 模型轉(zhuǎn)換為SavedModel
signature = tf.saved_model.signature_def_utils.predict_signature_def(
    inputs={"input_param": model.input}, outputs={"type": model.output})
builder = tf.saved_model.builder.SavedModelBuilder("/tmp/output_model_path/1/")
builder.add_meta_graph_and_variables(
    sess=K.get_session(),
    tags=[tf.saved_model.tag_constants.SERVING],
    signature_def_map={
        tf.saved_model.signature_constants.DEFAULT_SERVING_SIGNATURE_DEF_KEY:
            signature
    })
builder.save()

下載完成后，文件目錄樹為：

$ tree /tmp/resnet
/tmp/resnet
└── 1538687457
    ├── saved_model.pb
    └── variables
        ├── variables.data-00000-of-00001
        └── variables.index
部署模型

使用Docker部署模型服務(wù)：

$ docker pull tensorflow/serving
$ docker run -p 8500:8500 -p 8501:8501 --name tfserving_resnet 
--mount type=bind,source=/tmp/resnet,target=/models/resnet 
-e MODEL_NAME=resnet -t tensorflow/serving

其中，8500端口對于TensorFlow Serving提供的gRPC端口，8501為REST API服務(wù)端口。-e MODEL_NAME=resnet指出TensorFlow Serving需要加載的模型名稱，這里為resnet。上述命令輸出為

2019-03-04 02:52:26.610387: I tensorflow_serving/model_servers/server.cc:82] Building single TensorFlow model file config:  model_name: resnet model_base_path: /models/resnet
2019-03-04 02:52:26.618200: I tensorflow_serving/model_servers/server_core.cc:461] Adding/updating models.
2019-03-04 02:52:26.618628: I tensorflow_serving/model_servers/server_core.cc:558]  (Re-)adding model: resnet
2019-03-04 02:52:26.745813: I tensorflow_serving/core/basic_manager.cc:739] Successfully reserved resources to load servable {name: resnet version: 1538687457}
2019-03-04 02:52:26.745901: I tensorflow_serving/core/loader_harness.cc:66] Approving load for servable version {name: resnet version: 1538687457}
2019-03-04 02:52:26.745935: I tensorflow_serving/core/loader_harness.cc:74] Loading servable version {name: resnet version: 1538687457}
2019-03-04 02:52:26.747590: I external/org_tensorflow/tensorflow/contrib/session_bundle/bundle_shim.cc:363] Attempting to load native SavedModelBundle in bundle-shim from: /models/resnet/1538687457
2019-03-04 02:52:26.747705: I external/org_tensorflow/tensorflow/cc/saved_model/reader.cc:31] Reading SavedModel from: /models/resnet/1538687457
2019-03-04 02:52:26.795363: I external/org_tensorflow/tensorflow/cc/saved_model/reader.cc:54] Reading meta graph with tags { serve }
2019-03-04 02:52:26.828614: I external/org_tensorflow/tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2019-03-04 02:52:26.923902: I external/org_tensorflow/tensorflow/cc/saved_model/loader.cc:162] Restoring SavedModel bundle.
2019-03-04 02:52:28.098479: I external/org_tensorflow/tensorflow/cc/saved_model/loader.cc:138] Running MainOp with key saved_model_main_op on SavedModel bundle.
2019-03-04 02:52:28.144510: I external/org_tensorflow/tensorflow/cc/saved_model/loader.cc:259] SavedModel load for tags { serve }; Status: success. Took 1396689 microseconds.
2019-03-04 02:52:28.146646: I tensorflow_serving/servables/tensorflow/saved_model_warmup.cc:83] No warmup data file found at /models/resnet/1538687457/assets.extra/tf_serving_warmup_requests
2019-03-04 02:52:28.168063: I tensorflow_serving/core/loader_harness.cc:86] Successfully loaded servable version {name: resnet version: 1538687457}
2019-03-04 02:52:28.174902: I tensorflow_serving/model_servers/server.cc:286] Running gRPC ModelServer at 0.0.0.0:8500 ...
[warn] getaddrinfo: address family for nodename not supported
2019-03-04 02:52:28.186724: I tensorflow_serving/model_servers/server.cc:302] Exporting HTTP/REST API at:localhost:8501 ...
[evhttp_server.cc : 237] RAW: Entering the event loop ...

我們可以看到，TensorFlow Serving使用1538687457作為模型的版本號。我們使用curl命令來查看一下啟動的服務(wù)狀態(tài)，也可以看到提供服務(wù)的模型版本以及模型狀態(tài)。

$ curl http://localhost:8501/v1/models/resnet
{
 "model_version_status": [
  {
   "version": "1538687457",
   "state": "AVAILABLE",
   "status": {
    "error_code": "OK",
    "error_message": ""
   }
  }
 ]
}
查看模型輸入輸出

很多時候我們需要查看模型的輸出和輸出參數(shù)的具體形式，TensorFlow提供了一個saved_model_cli命令來查看模型的輸入和輸出參數(shù)：

$ saved_model_cli show --dir /tmp/resnet/1538687457/ --all

MetaGraphDef with tag-set: "serve" contains the following SignatureDefs:

signature_def["predict"]:
  The given SavedModel SignatureDef contains the following input(s):
    inputs["image_bytes"] tensor_info:
        dtype: DT_STRING
        shape: (-1)
        name: input_tensor:0
  The given SavedModel SignatureDef contains the following output(s):
    outputs["classes"] tensor_info:
        dtype: DT_INT64
        shape: (-1)
        name: ArgMax:0
    outputs["probabilities"] tensor_info:
        dtype: DT_FLOAT
        shape: (-1, 1001)
        name: softmax_tensor:0
  Method name is: tensorflow/serving/predict

signature_def["serving_default"]:
  The given SavedModel SignatureDef contains the following input(s):
    inputs["image_bytes"] tensor_info:
        dtype: DT_STRING
        shape: (-1)
        name: input_tensor:0
  The given SavedModel SignatureDef contains the following output(s):
    outputs["classes"] tensor_info:
        dtype: DT_INT64
        shape: (-1)
        name: ArgMax:0
    outputs["probabilities"] tensor_info:
        dtype: DT_FLOAT
        shape: (-1, 1001)
        name: softmax_tensor:0
  Method name is: tensorflow/serving/predict

注意到signature_def，inputs的名稱，類型和輸出，這些參數(shù)在接下來的模型預(yù)測請求中需要。

使用模型接口預(yù)測：REST和gRPC

TensorFlow Serving提供REST API和gRPC兩種請求方式，接下來將具體這兩種方式。

REST

我們下載一個客戶端腳本，這個腳本會下載一張貓的圖片，同時使用這張圖片來計算服務(wù)請求時間。

$ curl -o /tmp/resnet/resnet_client.py https://raw.githubusercontent.com/tensorflow/serving/master/tensorflow_serving/example/resnet_client.py

以下腳本使用requests庫來請求接口，使用圖片的base64編碼字符串作為請求內(nèi)容，返回圖片分類，并計算了平均處理時間。

from __future__ import print_function

import base64
import requests

# The server URL specifies the endpoint of your server running the ResNet
# model with the name "resnet" and using the predict interface.
SERVER_URL = "http://localhost:8501/v1/models/resnet:predict"

# The image URL is the location of the image we should send to the server
IMAGE_URL = "https://tensorflow.org/images/blogs/serving/cat.jpg"


def main():
  # Download the image
  dl_request = requests.get(IMAGE_URL, stream=True)
  dl_request.raise_for_status()

  # Compose a JSON Predict request (send JPEG image in base64).
  jpeg_bytes = base64.b64encode(dl_request.content).decode("utf-8")
  predict_request = "{"instances" : [{"b64": "%s"}]}" % jpeg_bytes

  # Send few requests to warm-up the model.
  for _ in range(3):
    response = requests.post(SERVER_URL, data=predict_request)
    response.raise_for_status()

  # Send few actual requests and report average latency.
  total_time = 0
  num_requests = 10
  for _ in range(num_requests):
    response = requests.post(SERVER_URL, data=predict_request)
    response.raise_for_status()
    total_time += response.elapsed.total_seconds()
    prediction = response.json()["predictions"][0]

  print("Prediction class: {}, avg latency: {} ms".format(
      prediction["classes"], (total_time*1000)/num_requests))


if __name__ == "__main__":
  main()

輸出結(jié)果為

$ python resnet_client.py
Prediction class: 286, avg latency: 210.12310000000002 ms
gRPC

讓我們下載另一個客戶端腳本，這個腳本使用gRPC作為服務(wù)，傳入圖片并獲取輸出結(jié)果。這個腳本需要安裝tensorflow-serving-api這個庫。

$ curl -o /tmp/resnet/resnet_client_grpc.py https://raw.githubusercontent.com/tensorflow/serving/master/tensorflow_serving/example/resnet_client_grpc.py
$ pip install tensorflow-serving-api

腳本內(nèi)容：

from __future__ import print_function

# This is a placeholder for a Google-internal import.

import grpc
import requests
import tensorflow as tf

from tensorflow_serving.apis import predict_pb2
from tensorflow_serving.apis import prediction_service_pb2_grpc

# The image URL is the location of the image we should send to the server
IMAGE_URL = "https://tensorflow.org/images/blogs/serving/cat.jpg"

tf.app.flags.DEFINE_string("server", "localhost:8500",
                           "PredictionService host:port")
tf.app.flags.DEFINE_string("image", "", "path to image in JPEG format")
FLAGS = tf.app.flags.FLAGS


def main(_):
  if FLAGS.image:
    with open(FLAGS.image, "rb") as f:
      data = f.read()
  else:
    # Download the image since we weren"t given one
    dl_request = requests.get(IMAGE_URL, stream=True)
    dl_request.raise_for_status()
    data = dl_request.content

  channel = grpc.insecure_channel(FLAGS.server)
  stub = prediction_service_pb2_grpc.PredictionServiceStub(channel)
  # Send request
  # See prediction_service.proto for gRPC request/response details.
  request = predict_pb2.PredictRequest()
  request.model_spec.name = "resnet"
  request.model_spec.signature_name = "serving_default"
  request.inputs["image_bytes"].CopyFrom(
      tf.contrib.util.make_tensor_proto(data, shape=[1]))
  result = stub.Predict(request, 10.0)  # 10 secs timeout
  print(result)


if __name__ == "__main__":
  tf.app.run()

輸出的結(jié)果可以看到圖片的分類，概率和使用的模型信息：

$ python resnet_client_grpc.py
outputs {
  key: "classes"
  value {
    dtype: DT_INT64
    tensor_shape {
      dim {
        size: 1
      }
    }
    int64_val: 286
  }
}
outputs {
  key: "probabilities"
  value {
    dtype: DT_FLOAT
    tensor_shape {
      dim {
        size: 1
      }
      dim {
        size: 1001
      }
    }
    float_val: 2.4162832232832443e-06
    float_val: 1.9012182974620373e-06
    float_val: 2.7247710022493266e-05
    float_val: 4.426385658007348e-07
    ...(中間省略)
    float_val: 1.4636580090154894e-05
    float_val: 5.812107133351674e-07
    float_val: 6.599806511076167e-05
    float_val: 0.0012952701654285192
  }
}
model_spec {
  name: "resnet"
  version {
    value: 1538687457
  }
  signature_name: "serving_default"
}
性能 通過編譯優(yōu)化的TensorFlow Serving二進(jìn)制來提高性能

TensorFlows serving有時會有輸出如下的日志：

Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA

TensorFlow Serving已發(fā)布Docker鏡像旨在盡可能多地使用CPU架構(gòu)，因此省略了一些優(yōu)化以最大限度地提高兼容性。如果你沒有看到此消息，則你的二進(jìn)制文件可能已針對你的CPU進(jìn)行了優(yōu)化。根據(jù)你的模型執(zhí)行的操作，這些優(yōu)化可能會對你的服務(wù)性能產(chǎn)生重大影響。幸運(yùn)的是，編譯優(yōu)化的TensorFlow Serving二進(jìn)制非常簡單。官方已經(jīng)提供了自動化腳本，分以下兩部進(jìn)行：

# 1. 編譯開發(fā)版本
$ docker build -t $USER/tensorflow-serving-devel -f Dockerfile.devel https://github.com/tensorflow/serving.git#:tensorflow_serving/tools/docker

# 2. 生產(chǎn)新的鏡像
$ docker build -t $USER/tensorflow-serving --build-arg TF_SERVING_BUILD_IMAGE=$USER/tensorflow-serving-devel https://github.com/tensorflow/serving.git#:tensorflow_serving/tools/docker

之后，使用新編譯的$USER/tensorflow-serving重新啟動服務(wù)即可。

總結(jié)

上面我們快速實(shí)踐了使用TensorFlow Serving和Docker部署機(jī)器學(xué)習(xí)服務(wù)的過程，可以看到，TensorFlow Serving提供了非常方便和高效的模型管理，配合Docker，可以快速搭建起機(jī)器學(xué)習(xí)服務(wù)。

參考

Serving ML Quickly with TensorFlow Serving and Docker

Train and serve a TensorFlow model with TensorFlow Serving

GitHub repo: qiwihui/blog
Follow me: @qiwihui


Site: QIWIHUI

文章版權(quán)歸作者所有，未經(jīng)允許請勿轉(zhuǎn)載,若此文章存在違規(guī)行為，您可以聯(lián)系管理員刪除。

轉(zhuǎn)載請注明本文地址：http://specialneedsforspecialkids.com/yun/27704.html

相關(guān)文章

  • TensorFlow在產(chǎn)品環(huán)境中運(yùn)行模型的實(shí)踐經(jīng)驗(yàn)總結(jié)

    摘要：它使用機(jī)器學(xué)習(xí)來解釋用戶提出的問題，并用相應(yīng)的知識庫文章來回應(yīng)。使用一類目前較先進(jìn)的機(jī)器學(xué)習(xí)算法來識別相關(guān)文章，也就是深度學(xué)習(xí)。接下來介紹一下我們在生產(chǎn)環(huán)境中配置模型的一些經(jīng)驗(yàn)。 我們?nèi)绾伍_始使用TensorFlow ?在Zendesk，我們開發(fā)了一系列機(jī)器學(xué)習(xí)產(chǎn)品，比如的自動答案（Automatic Answers）。它使用機(jī)器學(xué)習(xí)來解釋用戶提出的問題，并用相應(yīng)的知識庫文章來回應(yīng)。當(dāng)用戶有...

    stackfing 評論0 收藏0

  • 使用Kubernetes和TensorFlow Serving將神經(jīng)網(wǎng)絡(luò)鏡像分類進(jìn)行彈性擴(kuò)容

    摘要：大會以機(jī)器學(xué)習(xí)資料中心和云端安全為主要議題，為未來發(fā)展做戰(zhàn)略規(guī)劃。在年，谷歌開發(fā)了一個內(nèi)部深度學(xué)習(xí)基礎(chǔ)設(shè)施叫做，這個設(shè)施允許谷歌人創(chuàng)建更大的神經(jīng)網(wǎng)絡(luò)和擴(kuò)容實(shí)訓(xùn)成千上萬個核心。 導(dǎo)言 Google近日3月23－24日在美國舊金山舉辦首次谷歌云平臺（Google Cloud Platform） GCP NEXT大會，參會人數(shù)超過2000人。GCP NEXT大會以機(jī)器學(xué)習(xí)、資料中心和云端安全...

    Harpsichord1207 評論0 收藏0

  • 滴滴機(jī)器學(xué)習(xí)平臺架構(gòu)演進(jìn)

    摘要：滴滴機(jī)器學(xué)習(xí)平臺的治理思路主要是減少重復(fù)提高效率。本文將對滴滴的機(jī)器學(xué)習(xí)平臺進(jìn)行全面解讀，重點(diǎn)分享機(jī)器學(xué)習(xí)平臺不同階段所要解決的問題，以及解決問題的思路和技術(shù)方案。綜合和各自的利弊，滴滴機(jī)器學(xué)習(xí)平臺開始由架構(gòu)向建構(gòu)遷移。 前言：現(xiàn)在很多互聯(lián)網(wǎng)公司都有自己的機(jī)器學(xué)習(xí)平臺，冠以之名雖然形形色色，但就平臺所要解決的問題和技術(shù)選型基本還是大同小異。所謂大同是指大家所要處理的問題都相似，技術(shù)架構(gòu)...

    entner 評論0 收藏0

發(fā)表評論

0條評論

最新活動
閱讀需要支付1元查看
<