tensorflow serving使用

1、添加源

从官网下载，或者手动安装

!echo "deb http://storage.googleapis.com/tensorflow-serving-apt stable tensorflow-model-server tensorflow-model-server-universal" | tee /etc/apt/sources.list.d/tensorflow-serving.list && \
curl https://storage.googleapis.com/tensorflow-serving-apt/tensorflow-serving.release.pub.gpg | apt-key add -
!apt update

!apt-get install tensorflow-model-server

2、加载模型启动Tensorflow serving

几个重要参数：
rest_api_port: 服务的端口号
model_name: 取个模型名字
model_base_path: 存储模型的路径

# 模型路径
os.environ["MODEL_DIR"] = MODEL_DIR

%%bash --bg 
nohup tensorflow_model_server \
  --rest_api_port=8501 \
  --model_name=fashion_model \
  --model_base_path="${MODEL_DIR}" >server.log 2>&1

!tail server.log

向Tensorflow发起请求预估

# 构造请求内容
import json
data = json.dumps({"signature_name": "serving_default", "instances": test_images[0:3].tolist()})
print('发送的请求数据: {} ... {}'.format(data[:50], data[len(data)-52:]))

# 我们使用requests这个工具库发送POST请求到tensorflow serving的服务，并获取返回结果
!pip install -q requests

import requests
headers = {"content-type": "application/json"}
json_response = requests.post('http://localhost:8501/v1/models/fashion_model:predict', data=data, headers=headers)
predictions = json.loads(json_response.text)['predictions']

将predictions对应到class的名称即可