torchserve使用-网络端访问（三）

目录

1、通过http访问Using REST APIs

2、Using GRPC APIs through python client

---

通过脚本实现网络端访问。

1、通过http访问Using REST APIs

curl http://127.0.0.1:8080/predictions/densenet161 -T examples/image_classifier/kitten.jpg

实现相同功能：

import cv2
from PIL import Image
from io import BytesIO

def test():
    filename = "examples/image_classifier/kitten.jpg"
    image = Image.fromarray(cv2.imread(filename))
    image2bytes = BytesIO()
    image.save(image2bytes, format="PNG")
    image2bytes.seek(0)
    image_as_bytes = image2bytes.read()

    # Send the HTTP POST request to TorchServe
    import requests

    req = requests.post("http://127.0.0.1:8080/predictions/densenet161", data=image_as_bytes)
    if req.status_code == 200:
        res = req.json()
        print("ret:", res)

结果：

ret: {'tabby': 0.39039885997772217, 'tiger_cat': 0.36372312903404236,
 'Egyptian_cat': 0.21934737265110016, 'lynx': 0.021873705089092255,
  'snow_leopard': 0.0012753388145938516}

2、Using GRPC APIs through python client

Using GRPC APIs through python client

• Install grpc python dependencies :

pip install -U grpcio protobuf grpcio-tools

• Generate inference client using proto files

python -m grpc_tools.protoc --proto_path=frontend/server/src/main/resources/proto/ --python_out=ts_scripts --grpc_python_out=ts_scripts frontend/server/src/main/resources/proto/inference.proto frontend/server/src/main/resources/proto/management.proto

• Run inference using a sample client
• gRPC python client

python ts_scripts/torchserve_grpc_client.py infer densenet161 examples/image_classifier/kitten.jpg

https://github.com/pytorch/serve/blob/master/ts_scripts/torchserve_grpc_client.py

更多参数配置：

https://github.com/pytorch/serve/blob/v0.2.0/docs/batch_inference_with_ts.md

端口配置：

https://github.com/pytorch/serve/blob/v0.2.0/docs/configuration.md

上一节：torchserve使用-注册模型设置参数（二），下一节：torchserve使用-Torch Model archiver for TorchServe（四）