The goal of Paddle Serving is to provide high-performance, flexible and easy-to-use industrial-grade online inference services for machine learning developers and enterprises.Paddle Serving supports multiple protocols such as RESTful, gRPC, bRPC, and provides inference solutions under a variety of hardware and multiple operating system environments, and many famous pre-trained model examples. The core features are as follows:
- Integrate high-performance server-side inference engine paddle Inference and mobile-side engine paddle Lite. Models of other machine learning platforms (Caffe/TensorFlow/ONNX/PyTorch) can be migrated to paddle through [x2paddle](https://github.com/PaddlePaddle/X2Paddle).
- Integrate high-performance server-side inference engine [Paddle Inference](https://paddleinference.paddlepaddle.org.cn/product_introduction/inference_intro.html) and mobile-side engine [Paddle Lite](https://paddlelite.paddlepaddle.org.cn/introduction/tech_highlights.html). Models of other machine learning platforms (Caffe/TensorFlow/ONNX/PyTorch) can be migrated to paddle through [x2paddle](https://github.com/PaddlePaddle/X2Paddle).
- There are two frameworks, namely high-performance C++ Serving and high-easy-to-use Python pipeline. The C++ Serving is based on the bRPC network framework to create a high-throughput, low-latency inference service, and its performance indicators are ahead of competing products. The Python pipeline is based on the gRPC/gRPC-Gateway network framework and the Python language to build a highly easy-to-use and high-throughput inference service. How to choose which one please see [Techinical Selection](doc/Serving_Design_EN.md#21-design-selection).
- Support multiple [protocols](doc/C++_Serving/Inference_Protocols_CN.md) such as HTTP, gRPC, bRPC, and provide C++, Python, Java language SDK.
- Design and implement a high-performance inference service framework for asynchronous pipelines based on directed acyclic graph (DAG), with features such as multi-model combination, asynchronous scheduling, concurrent inference, dynamic batch, multi-card multi-stream inference, request cache, etc.
...
...
@@ -40,13 +40,17 @@ The goal of Paddle Serving is to provide high-performance, flexible and easy-to-
- Support service monitoring, provide prometheus-based performance statistics and port access
- Paper : [JiZhi: A Fast and Cost-Effective Model-As-A-Service System for
Web-Scale Online Inference at Baidu](https://arxiv.org/pdf/2106.01674.pdf)
...
...
@@ -67,6 +71,7 @@ This chapter guides you through the installation and deployment steps. It is str
-[Install Paddle Serving using docker](doc/Install_EN.md)
-[Build Paddle Serving from Source with Docker](doc/Compile_EN.md)
-[Install Paddle Serving on linux system](doc/Install_Linux_Env_CN.md)
-[Deploy Paddle Serving on Kubernetes(Chinese)](doc/Run_On_Kubernetes_CN.md)
-[Deploy Paddle Serving with Security gateway(Chinese)](doc/Serving_Auth_Docker_CN.md)
- Deploy on more hardwares[[ARM CPU、百度昆仑](doc/Run_On_XPU_EN.md)、[华为昇腾](doc/Run_On_NPU_CN.md)、[海光DCU](doc/Run_On_DCU_CN.md)、[Jetson](doc/Run_On_JETSON_CN.md)]
...
...
@@ -93,10 +98,11 @@ The first step is to call the model save interface to generate a model parameter
在`product_path`下运行下面的Python代码生产模型（运行前需要修改hadoop相关的参数），每隔 60 秒会产出 Boston 房价预测模型的打包文件`uci_housing.tar.gz`并上传至hdfs的`/`路径下，上传完毕后更新时间戳文件`donefile`并上传至hdfs的`/`路径下。
在`python/paddle_serving_server/server.py` 文件中仅添加`需要加载模型，执行推理预测的自定义的 C++ OP 类的类名`。例如 `GeneralReaderOp` 由于只是做一些简单的数据处理而不加载模型调用预测，故在上述的代码中需要添加，而不添加在下方的代码中。
AttributeError: 'Program' object has no attribute '_remove_training_info'
```
(3) CUDA 10.1及更高版本需要 TensorRT。安装 TensorRT 相关文件的脚本参考 [install_trt.sh](../tools/dockerfiles/build_scripts/install_trt.sh).
**A:** Paddle版本低，升级Paddle版本到2.2.x及以上
***
<aname="6"></a>
## 部署问题
#### Q: GPU环境运行Serving报错，GPU count is: 0。
#### Q: GPU 环境运行 Serving 报错，GPU count is: 0。
```
terminate called after throwing an instance of 'paddle::platform::EnforceNotMet'
...
...
@@ -261,34 +227,30 @@ InvalidArgumentError: Device id must be less than GPU count, but received id is:
[Hint: Expected id < GetCUDADeviceCount(), but received id:0 >= GetCUDADeviceCount():0.] at (/home/scmbuild/workspaces_cluster.dev/baidu.lib.paddlepaddle/baidu/lib/paddlepaddle/Paddle/paddle/fluid/platform/gpu_info.cc:211)
for most users, we do not need to read this section. But if you deploy your Paddle Serving on a machine without network, you will encounter a problem that the binary executable tar file cannot be downloaded. Therefore, here we give you all the download links for various environment.
In order to better optimize the performance, PipelineServing provides a timeline tool to monitor the time of each stage of the whole service.
### 1.2 Output profile information on server side
The server is controlled by the `use_profile` field in yaml:
```yaml
dag:
use_profile:true
```
After the function is enabled, the server will print the corresponding log information to the standard output in the process of prediction. In order to show the time consumption of each stage more intuitively, Analyst module is provided for further analysis and processing of log files.
The output of the server is first saved to a file. Taking `profile.txt` as an example, the script converts the time monitoring information in the log into JSON format and saves it to the `trace` file. The `trace` file can be visualized through the tracing function of Chrome browser.
```shell
from paddle_serving_server.pipeline import Analyst
import json
import sys
if __name__ =="__main__":
log_filename ="profile.txt"
trace_filename ="trace"
analyst = Analyst(log_filename)
analyst.save_trace(trace_filename)
```
Specific operation: open Chrome browser, input in the address bar `chrome://tracing/` , jump to the tracing page, click the load button, open the saved `trace` file, and then visualize the time information of each stage of the prediction service.
### 1.3 Output profile information on client side
The profile function can be enabled by setting `profile=True` in the `predict` interface on the client side.
After the function is enabled, the client will print the log information corresponding to the prediction to the standard output during the prediction process, and the subsequent analysis and processing are the same as that of the server.
### 1.4 Analytical methods
According to the time consumption of each stage in the pipeline.tracer log, the following formula is used to gradually analyze which stage is the main time consumption.
-[Performance Analysis And Optimization](Pipeline_Design_EN.md#6performance-analysis-and-optimization)
In many deep learning frameworks, Serving is usually used for the deployment of single model.but in the context of AI industrial, the end-to-end deep learning model can not solve all the problems at present. Usually, it is necessary to use multiple deep learning models to solve practical problems.However, the design of multi-model applications is complicated. In order to reduce the difficulty of development and maintenance, and to ensure the availability of services, serial or simple parallel methods are usually used. In general, the throughput only reaches the usable state and the GPU utilization rate is low.
Paddle Serving provides a user-friendly programming framework for multi-model composite services, Pipeline Serving, which aims to reduce the threshold of programming, improve resource utilization (especially GPU), and improve the prediction efficiency.
## 1.Architecture Design
The Server side is built based on <b>RPC Service</b> and <b>graph execution engine</b>. The relationship between them is shown in the following figure.
In order to meet the needs of different users, the RPC service starts one Web server and one RPC server at the same time, and can process 2 types of requests, RESTful API and gRPC.The gPRC gateway receives RESTful API requests and forwards requests to the gRPC server through the reverse proxy server; gRPC requests are received by the gRPC server, so the two types of requests are processed by the gRPC Service in a unified manner to ensure that the processing logic is consistent.
#### <b>1.1.1 Request and Respose of proto</b>
gRPC service and gRPC gateway service are generated with service.proto.
```proto
messageRequest{
repeatedstringkey=1;
repeatedstringvalue=2;
optionalstringname=3;
optionalstringmethod=4;
optionalint64logid=5;
optionalstringclientip=6;
};
messageResponse{
optionalint32err_no=1;
optionalstringerr_msg=2;
repeatedstringkey=3;
repeatedstringvalue=4;
};
```
The `key` and `value` in the Request are paired string arrays. The `name` and `method` correspond to the URL of the RESTful API://{ip}:{port}/{name}/{method}.The `logid` and `clientip` are convenient for users to connect service-level requests and customize strategies.
In Response, `err_no` and `err_msg` express the correctness and error information of the processing result, and `key` and `value` are the returned results.
### 1.2 Graph Execution Engine
The graph execution engine consists of OPs and Channels, and the connected OPs share one Channel.
- Channel can be understood as a buffer queue. Each OP accepts only one Channel input and multiply Channel outputs (each output is the same); a Channel can contain outputs from multiple OPs, and data from the same Channel can be used as input for multiple OPs.
- Users only need to define relationships between OPs. Graph engine will analyze the dependencies of the entire graph and declaring Channels at the compile time.
- After Request data enters the graph execution engine service, the graph engine will generator an Request ID, and Reponse is returned through corresponding Request ID.
- For cases where large data needs to be transferred between OPs, consider RAM DB external memory for global storage and data transfer by passing index keys in Channel.
- The default function of a single OP is to access a single Paddle Serving Service based on the input Channel data and put the result into the output Channel.
- OP supports user customization, including preprocess, process, postprocess functions that can be inherited and implemented by the user.
- OP can set the number of concurrencies to increase the number of concurrencies processed.
- OP can obtain data from multiple different RPC requests for Auto-Batching.
- OP can be started by a thread or process.
#### <b>1.2.2 Channel Design</b>
- Channel is the data structure for sharing data between OPs, responsible for sharing data or sharing data status information.
- Outputs from multiple OPs can be stored in the same Channel, and data from the same Channel can be used by multiple OPs.
- The following illustration shows the design of Channel in the graph execution engine, using input buffer and output buffer to align data between multiple OP inputs and multiple OP outputs, with a queue in the middle to buffer.
- Prediction type (client_type) of Op has 3 types, brpc, grpc and local_predictor
- brpc: Using bRPC Client to interact with remote Serving by network, performance is better than grpc.
- grpc: Using gRPC Client to interact with remote Serving by network, cross-platform deployment supported.
- local_predictor: Load the model and predict in the local service without interacting with the network. Support multi-card deployment, and TensorRT prediction.
- Selection:
- Time cost(lower is better): local_predict < brpc <= grpc
- Microservice: Split the brpc or grpc model into independent services, simplify development and deployment complexity, and improve resource utilization
#### <b>1.2.4 Extreme Case Consideration</b>
-`Request timeout`
The entire graph execution engine may time out at every step. The graph execution engine controls the time out by setting `timeout` value. Requests that time out at any step will return a timeout response.
-`Channel stores too much data`
Channels may store too much data, causing copy time to be too high. Graph execution engines can store OP calculation results in external memory, such as high-speed memory KV systems.
-`Whether input buffers and output buffers in Channel will increase indefinitely`
- It will not increase indefinitely. The input to the entire graph execution engine is placed inside a Channel's internal queue, directly acting as a traffic control buffer queue for the entire service.
- For input buffer, adjust the number of concurrencies of OP1 and OP2 according to the amount of computation, so that the number of input buffers from each input OP is relatively balanced. (The length of the input buffer depends on the speed at which each item in the internal queue is ready)
- For output buffer, you can use a similar process as input buffer, which adjusts the concurrency of OP3 and OP4 to control the buffer length of output buffer. (The length of the output buffer depends on the speed at which downstream OPs obtain data from the output buffer)
- The amount of data in the Channel will not exceed `worker_num` of gRPC, that is, it will not exceed the thread pool size.
***
## 2.Detailed Design
For the design and implementation of Pipeline, first introduce PipelineServer, OP, pre- and post-processing of rewriting OP, and finally introduce the secondary development method of specific OP (RequestOp and ResponseOp).
### 2.1 PipelineServer Definition
PipelineServer encapsulates the RPC runtime layer and graph engine execution. All Pipeline services must first instantiate the PipelineServer example, then set up two core steps, set response op and load configuration information, and finally call run_server to start the service. The code example is as follows:
```python
server=PipelineServer()
server.set_response_op(response_op)
server.prepare_server(config_yml_path)
#server.prepare_pipeline_config(config_yml_path)
server.run_server()
```
The core interface of PipelineServer:
-`set_response_op`: setting response_op will initialize the Channel according to the topological relationship of each OP and build a calculation graph.
-`prepare_server`: load configuration information, and start remote Serving service, suitable for calling remote remote reasoning service.
-`prepare_pipeline_config`: only load configuration information, applicable to local_prdict
| name | (str) String used to identify the OP type, which must be globally unique. |
| input_ops | (list) A list of all previous OPs of the current Op. |
| server_endpoints | (list) List of endpoints for remote Paddle Serving Service. If this parameter is not set,it is considered as local_precditor mode, and the configuration is read from local_service_conf |
| fetch_list | (list) List of fetch variable names for remote Paddle Serving Service. |
| client_config | (str) The path of the client configuration file corresponding to the Paddle Serving Service. |
| client_type | （str)brpc, grpc or local_predictor. local_predictor does not start the Serving service, in-process prediction|
| concurrency | (int) The number of concurrent OPs. |
| timeout | (int) The timeout time of the process operation, in ms. If the value is less than zero, no timeout is considered. |
| retry | (int) Timeout number of retries. When the value is 1, no retries are made. |
| batch_size | (int) The expected batch_size of Auto-Batching, since building batches may time out, the actual batch_size may be less than the set value. |
| auto_batching_timeout | (float) Timeout for building batches of Auto-Batching (the unit is ms). When batch_size> 1, auto_batching_timeout should be set, otherwise the waiting will be blocked when the number of requests is insufficient for batch_size|
| local_service_handler | (object) local predictor handler，assigned by Op init() input parameters or created in Op init()|
| def preprocess(self, input_dicts) | Process the data obtained from the channel, and the processed data will be used as the input of the **process** function. (This function handles a **sample**) |
| def process(self, feed_dict_list, typical_logid) | The RPC prediction process is based on the Paddle Serving Client, and the processed data will be used as the input of the **postprocess** function. (This function handles a **batch**) |
| def postprocess(self, input_dicts, fetch_dict) | After processing the prediction results, the processed data will be put into the subsequent Channel to be obtained by the subsequent OP. (This function handles a **sample**) |
| def init_op(self) | Used to load resources (such as word dictionary). |
| self.concurrency_idx | Concurrency index of current process(not thread) (different kinds of OP are calculated separately). |
In a running cycle, OP will execute three operations: preprocess, process, and postprocess (when the `server_endpoints` parameter is not set, the process operation is not executed). Users can rewrite these three functions. The default implementation is as follows:
```python
defpreprocess(self,input_dicts):
# multiple previous Op
iflen(input_dicts)!=1:
raiseNotImplementedError(
'this Op has multiple previous inputs. Please override this func.'
The parameter of **preprocess** is the data `input_dicts` in the previous Channel. This variable (as a **sample**) is a dictionary with the name of the previous OP as key and the output of the corresponding OP as value.
The parameter of **process** is the input variable `fetch_dict_list` (a list of the return value of the preprocess function) of the Paddle Serving Client prediction interface. This variable (as a **batch**) is a list of dictionaries with feed_name as the key and the data in the ndarray format as the value. `typical_logid` is used as the logid that penetrates to PaddleServingService.
The parameters of **postprocess** are `input_dicts` and `fetch_dict`. `input_dicts` is consistent with the parameter of preprocess, and `fetch_dict` (as a **sample**) is a sample of the return batch of the process function (if process is not executed, this value is the return value of preprocess).
Users can also rewrite the **init_op** function to load some custom resources (such as word dictionary). The default implementation is as follows:
```python
definit_op(self):
pass
```
It should be **noted** that in the threaded version of OP, each OP will only call this function once, so the loaded resources must be thread safe.
### 2.4 RequestOp Definition and Secondary Development Interface
RequestOp is used to process RPC data received by Pipeline Server, and the processed data will be added to the graph execution engine. Its constructor is as follows:
The default implementation of **unpack_request_package** is to make the key and value in RPC request into a dictionary.When the default RequestOp cannot meet the parameter parsing requirements, you can customize the request parameter parsing method by rewriting the following two interfaces.The return value is required to be a dictionary type.
| def init_op(self) | It is used to load resources (such as dictionaries), and is consistent with general OP. |
| def unpack_request_package(self, request) | Process received RPC data. |
### 2.5 ResponseOp Definition and Secondary Development Interface
ResponseOp is used to process the prediction results of the graph execution engine. The processed data will be used as the RPC return value of Pipeline Server. Its constructor is as follows:
The default implementation of **pack_response_package** is to convert the dictionary of prediction results into key and value in RPC response.When the default ResponseOp cannot meet the requirements of the result return format, you can customize the return package packaging method by rewriting the following two interfaces.
Here, we build a simple imdb model enable example to show how to use Pipeline Serving. The relevant code can be found in the `Serving/examples/Pipeline/imdb_model_ensemble` folder. The Server-side structure in the example is shown in the following figure:
Five types of files are needed, of which model files, configuration files, and server code are the three necessary files for building a Pipeline service. Test client and test data set are prepared for testing.
- model files
- configure files(config.yml)
- service level: Service port, thread number, service timeout, retry, etc.
- DAG level: Resource type, enable Trace, performance profile, etc.
- OP level: Model path, concurrency, client type, device type, automatic batching, etc.
- Server files(web_server.py)
- service level: Define service name, read configuration file, start service, etc.
- DAG level: Topological relationship between OPs.
- OP level: Rewrite preprocess and postprocess of OP.
Users can customize the error code according to the business, inherit ProductErrCode, and return it in the return list in Op's preprocess or postprocess. The next stage of processing will skip the post OP processing based on the custom error code.
```python
classProductErrCode(enum.Enum):
"""
ProductErrCode is a base class for recording business error code.
product developers inherit this class and extend more error codes.
"""
pass
```
### 4.2 Skip process stage
The 2rd result of the result list returned by preprocess is `is_skip_process=True`, indicating whether to skip the process stage of the current OP and directly enter the postprocess processing
```python
defpreprocess(self,input_dicts,data_id,log_id):
"""
In preprocess stage, assembling data for process stage. users can
override this function for model feed features.
Args:
input_dicts: input data to be preprocessed
data_id: inner unique id
log_id: global unique id for RTT
Return:
input_dict: data for process stage
is_skip_process: skip process stage or not, False default
"Failed to run preprocess: this Op has multiple previous "
"inputs. Please override this func."))
os._exit(-1)
(_,input_dict),=input_dicts.items()
returninput_dict,False,None,""
```
### 4.3 Custom proto Request and Response
When the default proto structure does not meet the business requirements, at the same time, the Request and Response message structures of the proto in the following two files remain the same.
> pipeline/gateway/proto/gateway.proto
> pipeline/proto/pipeline_service.proto
Recompile Serving Server again.
### 4.4 Custom URL
The grpc gateway processes post requests. The default `method` is `prediction`, for example: 127.0.0.1:8080/ocr/prediction. Users can customize the name and method, and can seamlessly switch services with existing URLs.
```proto
servicePipelineService{
rpcinference(Request)returns(Response){
option(google.api.http)={
post:"/{name=*}/{method=*}"
body:"*"
};
}
};
```
### 4.5 Batch predictor
Pipeline supports batch predictor, and GPU utilization can be improved by increasing the batch size. Pipeline Serving supports 3 Pipeline Serving supports 3 batch forms and applicable scenarios are as follows:
- case 1: An inference request contains batch data (batch)
- The data is of fixed length, the batch is variable, and the data is converted into BCHW format
- The data length is variable. In the pre-processing, a single piece of data is padding converted into a fixed length
- case 2: Split the batch data of a inference request into multiple small pieces of data (mini-batch)
- Since padding will be aligned at the longest shape, when there is a "extremely large" shape size data in a batch of data, the padding is very large.
- Specify the size of a block to reduce the scope of the "extremely large" size data
- case 3: Merge multiple requests for one batch(auto-batching)
- Inference time is significantly longer than preprocess and postprocess. Merge multiple request data and inference at one time will increase throughput and GPU utilization.
- The shape of the data of multiple requests is required to be consistent
| mini-batch | the return type of preprocess is list，refer to the preprocess of RecOp in OCR example|
| auto-batching | set batch_size and auto_batching_timeout in config.yml |
### 4.6 Single-machine and multi-card inference
Single-machine multi-card inference can be abstracted into M OP processes bound to N GPU cards. It is related to the configuration of three parameters in config.yml. First, select the process mode, the number of concurrent processes is the number of processes, and devices is the GPU card ID.The binding method is to traverse the GPU card ID when the process starts, for example, start 7 OP processes, set devices:0,1,2 in config.yml, then the first, fourth, and seventh started processes are bound to the 0 card, and the second , 4 started processes are bound to 1 card, 3 and 6 processes are bound to card 2.
In addition to supporting CPU and GPU, Pipeline also supports the deployment of a variety of heterogeneous hardware. It consists of device_type and devices in config.yml. Use device_type to specify the type first, and judge according to devices when it is vacant. The device_type is described as follows:
Pipeline Serving supports low-precision inference. The precision types supported by CPU, GPU and TensoRT are shown in the figure below:
- CPU
- fp32(default)
- fp16
- bf16(mkldnn)
- GPU
- fp32(default)
- fp16
- int8
- Tensor RT
- fp32(default)
- fp16
- int8
Reference the example [simple_web_service](../../examples/Pipeline/simple_web_service).
***
## 5.Log Tracing
Pipeline service logs are under the `PipelineServingLogs` directory of the current directory. There are 3 types of logs, namely `pipeline.log`, `pipeline.log.wf`, and `pipeline.tracer`.
- pipeline.log : Record debug & info level log
- pipeline.log.wf : Record warning & error level log
- pipeline.tracer : Statistics the time-consuming and channel accumulation information in each stage
When an exception occurs in the service, the error message will be recorded in the file `pipeline.log.wf`. Printing the tracer log requires adding the tracer configuration in the DAG property of `config.yml`.
### 5.1 Log uniquely id
There are two kinds of IDs in the pipeline for concatenating requests, `data_id` and `log_id` respectively. The difference between the two is as follows:
-`data_id`: The self-incrementing ID generated by the pipeline framework, marking the unique identification of the request.
-`log_id`: The identifier passed in by the upstream module tracks the serial relationship between multiple services. Since users may not pass in or guarantee uniqueness, it cannot be used as a unique identifier.
The log printed by the Pipeline framework will carry both data_id and log_id. After auto-batching is turned on, the first `data_id` in the batch will be used to mark the whole batch, and the framework will print all data_ids in the batch in a log.
### 5.2 Log Rotating
Log module of Pipeline Serving is defined in file `logger.py`.`logging.handlers.RotatingFileHandler` is used to support the rotation of disk log files. Set `maxBytes` and `backupCount` according to different file levels and daily quality. When the predetermined size is about to be exceeded, the old file will be closed and a new file will be opened for output.
In order to better optimize the performance, PipelineServing provides a timeline tool to monitor the time of each stage of the whole service.
### 6.2 Output profile information on server side
The server is controlled by the `use_profile` field in yaml:
```yaml
dag:
use_profile:true
```
After the function is enabled, the server will print the corresponding log information to the standard output in the process of prediction. In order to show the time consumption of each stage more intuitively, Analyst module is provided for further analysis and processing of log files.
The output of the server is first saved to a file. Taking `profile.txt` as an example, the script converts the time monitoring information in the log into JSON format and saves it to the `trace` file. The `trace` file can be visualized through the tracing function of Chrome browser.
```shell
from paddle_serving_server.pipeline import Analyst
import json
import sys
if __name__ =="__main__":
log_filename ="profile.txt"
trace_filename ="trace"
analyst = Analyst(log_filename)
analyst.save_trace(trace_filename)
```
Specific operation: open Chrome browser, input in the address bar `chrome://tracing/` , jump to the tracing page, click the load button, open the saved `trace` file, and then visualize the time information of each stage of the prediction service.
### 6.3 Output profile information on client side
The profile function can be enabled by setting `profile=True` in the `predict` interface on the client side.
After the function is enabled, the client will print the log information corresponding to the prediction to the standard output during the prediction process, and the subsequent analysis and processing are the same as that of the server.
### 6.4 Analytical methods
According to the time consumption of each stage in the pipeline.tracer log, the following formula is used to gradually analyze which stage is the main time consumption.
程序会分别运行 cpu 和 gpu 示例。运行成功则打印 `Pipeline cpu environment running success
` 和 `Pipeline gpu environment running success`。
```
/usr/local/lib/python3.7/runpy.py:125: RuntimeWarning: 'paddle_serving_server.serve' found in sys.modules after import of package 'paddle_serving_server', but prior to execution of 'paddle_serving_server.serve'; this may result in unpredictable behaviour
warn(RuntimeWarning(msg))
Welcome to the check env shell.Type help to list commands.
C++ gpu environment running failure, if you need this environment, please refer to https://github.com/PaddlePaddle/Serving/blob/develop/doc/Install_CN.md
Traceback (most recent call last):
File "/usr/local/lib/python3.7/runpy.py", line 193, in _run_module_as_main
"__main__", mod_spec)
File "/usr/local/lib/python3.7/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/usr/local/lib/python3.7/site-packages/paddle_serving_server/serve.py", line 541, in <module>
Check_Env_Shell().cmdloop()
File "/usr/local/lib/python3.7/cmd.py", line 138, in cmdloop
stop = self.onecmd(line)
File "/usr/local/lib/python3.7/cmd.py", line 217, in onecmd
return func(arg)
File "/usr/local/lib/python3.7/site-packages/paddle_serving_server/serve.py", line 501, in do_check_all
check_env("all")
File "/usr/local/lib/python3.7/site-packages/paddle_serving_server/env_check/run.py", line 94, in check_env
在许多深度学习框架中，模型服务化部署通常用于单模型的一键部署。但在 AI 工业大生产的背景下，端到端的单一深度学习模型不能解决复杂问题，多个深度学习模型组合使用是解决现实复杂问题的常规手段，如文字识别 OCR 服务至少需要检测和识别2种模型；视频理解服务一般需要视频抽帧、切词、音频处理、分类等多种模型组合实现。当前，通用多模型组合服务的设计和实现是非常复杂的，既要能实现复杂的模型拓扑关系，又要保证服务的高并发、高可用和易于开发和维护等。
