未验证 提交 66fdccea 编写于 作者: M MRXLT 提交者: GitHub

Merge pull request #670 from MRXLT/0.3.0-cherry

[cherry pick to 0.3.0]
......@@ -61,9 +61,9 @@ Packages of Paddle Serving support Centos 6/7 and Ubuntu 16/18, or you can use H
<h3 align="center">Chinese Word Segmentation</h4>
``` shell
> python -m paddle_serving_app.package -get_model lac
> python -m paddle_serving_app.package --get_model lac
> tar -xzf lac.tar.gz
> python lac_web_service.py 9292 &
> python lac_web_service.py lac_model/ lac_workdir 9393 &
> curl -H "Content-Type:application/json" -X POST -d '{"feed":[{"words": "我爱北京天安门"}], "fetch":["word_seg"]}' http://127.0.0.1:9393/lac/prediction
{"result":[{"word_seg":"我|爱|北京|天安门"}]}
```
......@@ -77,7 +77,7 @@ Packages of Paddle Serving support Centos 6/7 and Ubuntu 16/18, or you can use H
<p>
``` shell
> python -m paddle_serving_app.package -get_model resnet_v2_50_imagenet
> python -m paddle_serving_app.package --get_model resnet_v2_50_imagenet
> tar -xzf resnet_v2_50_imagenet.tar.gz
> python resnet50_imagenet_classify.py resnet50_serving_model &
> curl -H "Content-Type:application/json" -X POST -d '{"feed":[{"image": "https://paddle-serving.bj.bcebos.com/imagenet-example/daisy.jpg"}], "fetch": ["score"]}' http://127.0.0.1:9292/image/prediction
......@@ -111,9 +111,9 @@ python -m paddle_serving_server.serve --model uci_housing_model --thread 10 --po
| `port` | int | `9292` | Exposed port of current service to users|
| `name` | str | `""` | Service name, can be used to generate HTTP request url |
| `model` | str | `""` | Path of paddle model directory to be served |
| `mem_optim` | bool | `False` | Enable memory / graphic memory optimization |
| `ir_optim` | bool | `False` | Enable analysis and optimization of calculation graph |
| `use_mkl` (Only for cpu version) | bool | `False` | Run inference with MKL |
| `mem_optim` | - | - | Enable memory / graphic memory optimization |
| `ir_optim` | - | - | Enable analysis and optimization of calculation graph |
| `use_mkl` (Only for cpu version) | - | - | Run inference with MKL |
Here, we use `curl` to send a HTTP POST request to the service we just started. Users can use any python library to send HTTP POST as well, e.g, [requests](https://requests.readthedocs.io/en/master/).
</center>
......@@ -170,13 +170,13 @@ Here, `client.predict` function has two arguments. `feed` is a `python dict` wit
### About Efficiency
- [How to profile Paddle Serving latency?](python/examples/util)
- [How to optimize performance?(Chinese)](doc/PERFORMANCE_OPTIM_CN.md)
- [How to optimize performance?](doc/PERFORMANCE_OPTIM.md)
- [Deploy multi-services on one GPU(Chinese)](doc/MULTI_SERVICE_ON_ONE_GPU_CN.md)
- [CPU Benchmarks(Chinese)](doc/BENCHMARKING.md)
- [GPU Benchmarks(Chinese)](doc/GPU_BENCHMARKING.md)
### FAQ
- [FAQ(Chinese)](doc/deprecated/FAQ.md)
- [FAQ(Chinese)](doc/FAQ.md)
### Design
......
......@@ -62,9 +62,9 @@ Paddle Serving安装包支持Centos 6/7和Ubuntu 16/18,或者您可以使用HT
<h3 align="center">中文分词</h4>
``` shell
> python -m paddle_serving_app.package -get_model lac
> python -m paddle_serving_app.package --get_model lac
> tar -xzf lac.tar.gz
> python lac_web_service.py 9292 &
> python lac_web_service.py lac_model/ lac_workdir 9393 &
> curl -H "Content-Type:application/json" -X POST -d '{"feed":[{"words": "我爱北京天安门"}], "fetch":["word_seg"]}' http://127.0.0.1:9393/lac/prediction
{"result":[{"word_seg":"我|爱|北京|天安门"}]}
```
......@@ -78,7 +78,7 @@ Paddle Serving安装包支持Centos 6/7和Ubuntu 16/18,或者您可以使用HT
<p>
``` shell
> python -m paddle_serving_app.package -get_model resnet_v2_50_imagenet
> python -m paddle_serving_app.package --get_model resnet_v2_50_imagenet
> tar -xzf resnet_v2_50_imagenet.tar.gz
> python resnet50_imagenet_classify.py resnet50_serving_model &
> curl -H "Content-Type:application/json" -X POST -d '{"feed":[{"image": "https://paddle-serving.bj.bcebos.com/imagenet-example/daisy.jpg"}], "fetch": ["score"]}' http://127.0.0.1:9292/image/prediction
......@@ -115,9 +115,9 @@ python -m paddle_serving_server.serve --model uci_housing_model --thread 10 --po
| `port` | int | `9292` | Exposed port of current service to users|
| `name` | str | `""` | Service name, can be used to generate HTTP request url |
| `model` | str | `""` | Path of paddle model directory to be served |
| `mem_optim` | bool | `False` | Enable memory optimization |
| `ir_optim` | bool | `False` | Enable analysis and optimization of calculation graph |
| `use_mkl` (Only for cpu version) | bool | `False` | Run inference with MKL |
| `mem_optim` | - | - | Enable memory optimization |
| `ir_optim` | - | - | Enable analysis and optimization of calculation graph |
| `use_mkl` (Only for cpu version) | - | - | Run inference with MKL |
我们使用 `curl` 命令来发送HTTP POST请求给刚刚启动的服务。用户也可以调用python库来发送HTTP POST请求,请参考英文文档 [requests](https://requests.readthedocs.io/en/master/)。
</center>
......@@ -181,7 +181,7 @@ print(fetch_map)
- [GPU版Benchmarks](doc/GPU_BENCHMARKING.md)
### FAQ
- [常见问答](doc/deprecated/FAQ.md)
- [常见问答](doc/FAQ.md)
### 设计文档
- [Paddle Serving设计文档](doc/DESIGN_DOC_CN.md)
......
......@@ -59,7 +59,7 @@ the script of client side bert_client.py is as follow:
import os
import sys
from paddle_serving_client import Client
from paddle_serving_app import ChineseBertReader
from paddle_serving_app.reader import ChineseBertReader
reader = ChineseBertReader()
fetch = ["pooled_output"]
......
......@@ -52,7 +52,7 @@ pip install paddle_serving_app
``` python
import sys
from paddle_serving_client import Client
from paddle_serving_app import ChineseBertReader
from paddle_serving_app.reader import ChineseBertReader
reader = ChineseBertReader()
fetch = ["pooled_output"]
......
# FAQ
- Q:如何调整RPC服务的等待时间,避免超时?
A:使用set_rpc_timeout_ms设置更长的等待时间,单位为毫秒,默认时间为20秒。
示例:
```
from paddle_serving_client import Client
client = Client()
client.load_client_config(sys.argv[1])
client.set_rpc_timeout_ms(100000)
client.connect(["127.0.0.1:9393"])
```
......@@ -99,7 +99,7 @@ func main() {
### 基于IMDB测试集的预测
```python
go run imdb_client.go serving_client_conf / serving_client_conf.stream.prototxt test.data> result
go run imdb_client.go serving_client_conf/serving_client_conf.stream.prototxt test.data> result
```
### 计算精度
......
......@@ -16,5 +16,5 @@ Parameters for performance optimization:
| Parameters | Type | Default | Description |
| ---------- | ---- | ------- | ------------------------------------------------------------ |
| mem_optim | bool | False | Enable memory / graphic memory optimization |
| ir_optim | bool | Fasle | Enable analysis and optimization of calculation graph,including OP fusion, etc |
| mem_optim | - | - | Enable memory / graphic memory optimization |
| ir_optim | - | - | Enable analysis and optimization of calculation graph,including OP fusion, etc |
......@@ -16,5 +16,5 @@
| 参数 | 类型 | 默认值 | 含义 |
| --------- | ---- | ------ | -------------------------------- |
| mem_optim | bool | False | 开启内存/显存优化 |
| ir_optim | bool | Fasle | 开启计算图分析优化,包括OP融合等 |
| mem_optim | - | - | 开启内存/显存优化 |
| ir_optim | - | - | 开启计算图分析优化,包括OP融合等 |
......@@ -40,10 +40,14 @@ def parse_args(): # pylint: disable=doc-string-missing
parser.add_argument(
"--device", type=str, default="cpu", help="Type of device")
parser.add_argument(
"--mem_optim", type=bool, default=False, help="Memory optimize")
"--mem_optim",
default=False,
action="store_true",
help="Memory optimize")
parser.add_argument(
"--ir_optim", type=bool, default=False, help="Graph optimize")
parser.add_argument("--use_mkl", type=bool, default=False, help="Use MKL")
"--ir_optim", default=False, action="store_true", help="Graph optimize")
parser.add_argument(
"--use_mkl", default=False, action="store_true", help="Use MKL")
parser.add_argument(
"--max_body_size",
type=int,
......
......@@ -47,9 +47,12 @@ def serve_args():
parser.add_argument(
"--name", type=str, default="None", help="Default service name")
parser.add_argument(
"--mem_optim", type=bool, default=False, help="Memory optimize")
"--mem_optim",
default=False,
action="store_true",
help="Memory optimize")
parser.add_argument(
"--ir_optim", type=bool, default=False, help="Graph optimize")
"--ir_optim", default=False, action="store_true", help="Graph optimize")
parser.add_argument(
"--max_body_size",
type=int,
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册