[Serving] add ppdet serving example (#641)

* serving support ppdet

* Update README.md

update ppadet/README

examples/vision/classification/paddleclas/serving/README.md

@@ -13,7 +13,7 @@ tar -xvf ResNet50_vd_infer.tgz
 wget https://gitee.com/paddlepaddle/PaddleClas/raw/release/2.4/deploy/images/ImageNet/ILSVRC2012_val_00000010.jpeg
 
 # 将配置文件放入预处理目录
-mv ResNet50_vd_infer/inference_cls.yaml models/preprocess/1/
+mv ResNet50_vd_infer/inference_cls.yaml models/preprocess/1/inference_cls.yaml
 
 # 将模型放入 models/runtime/1目录下, 并重命名为model.pdmodel和model.pdiparams
 mv ResNet50_vd_infer/inference.pdmodel models/runtime/1/model.pdmodel

examples/vision/detection/paddledetection/README.md

@@ -47,3 +47,4 @@
 
 - [Python部署](python)
 - [C++部署](cpp)
+- [服务化部署](serving)

examples/vision/detection/paddledetection/serving/models/postprocess/1/model.py

@@ -0,0 +1,110 @@
+# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import json
+import numpy as np
+import time
+
+import fastdeploy as fd
+
+# triton_python_backend_utils is available in every Triton Python model. You
+# need to use this module to create inference requests and responses. It also
+# contains some utility functions for extracting information from model_config
+# and converting Triton input/output types to numpy types.
+import triton_python_backend_utils as pb_utils
+
+
+class TritonPythonModel:
+    """Your Python model must use the same class name. Every Python model
+    that is created must have "TritonPythonModel" as the class name.
+    """
+
+    def initialize(self, args):
+        """`initialize` is called only once when the model is being loaded.
+        Implementing `initialize` function is optional. This function allows
+        the model to intialize any state associated with this model.
+        Parameters
+        ----------
+        args : dict
+          Both keys and values are strings. The dictionary keys and values are:
+          * model_config: A JSON string containing the model configuration
+          * model_instance_kind: A string containing model instance kind
+          * model_instance_device_id: A string containing model instance device ID
+          * model_repository: Model repository path
+          * model_version: Model version
+          * model_name: Model name
+        """
+        # You must parse model_config. JSON string is not parsed here
+        self.model_config = json.loads(args['model_config'])
+        print("model_config:", self.model_config)
+
+        self.input_names = []
+        for input_config in self.model_config["input"]:
+            self.input_names.append(input_config["name"])
+        print("postprocess input names:", self.input_names)
+
+        self.output_names = []
+        self.output_dtype = []
+        for output_config in self.model_config["output"]:
+            self.output_names.append(output_config["name"])
+            dtype = pb_utils.triton_string_to_numpy(output_config["data_type"])
+            self.output_dtype.append(dtype)
+        print("postprocess output names:", self.output_names)
+
+        self.postprocess_ = fd.vision.detection.PaddleDetPostprocessor()
+
+    def execute(self, requests):
+        """`execute` must be implemented in every Python model. `execute`
+        function receives a list of pb_utils.InferenceRequest as the only
+        argument. This function is called when an inference is requested
+        for this model. Depending on the batching configuration (e.g. Dynamic
+        Batching) used, `requests` may contain multiple requests. Every
+        Python model, must create one pb_utils.InferenceResponse for every
+        pb_utils.InferenceRequest in `requests`. If there is an error, you can
+        set the error argument when creating a pb_utils.InferenceResponse.
+        Parameters
+        ----------
+        requests : list
+          A list of pb_utils.InferenceRequest
+        Returns
+        -------
+        list
+          A list of pb_utils.InferenceResponse. The length of this list must
+          be the same as `requests`
+        """
+        responses = []
+        for request in requests:
+            infer_outputs = []
+            for name in self.input_names:
+                infer_output = pb_utils.get_input_tensor_by_name(request, name)
+                if infer_output:
+                    infer_output = infer_output.as_numpy()
+                    infer_outputs.append(infer_output)
+
+            results = self.postprocess_.run(infer_outputs)
+            r_str = fd.vision.utils.fd_result_to_json(results)
+
+            r_np = np.array(r_str, dtype=np.object)
+            out_tensor = pb_utils.Tensor(self.output_names[0], r_np)
+            inference_response = pb_utils.InferenceResponse(
+                output_tensors=[out_tensor, ])
+            responses.append(inference_response)
+        return responses
+
+    def finalize(self):
+        """`finalize` is called only once when the model is being unloaded.
+        Implementing `finalize` function is optional. This function allows
+        the model to perform any necessary clean ups before exit.
+        """
+        print('Cleaning up...')

examples/vision/detection/paddledetection/serving/models/postprocess/config.pbtxt

@@ -0,0 +1,30 @@
+name: "postprocess"
+backend: "python"
+
+input [
+  {
+    name: "post_input1"
+    data_type: TYPE_FP32
+    dims: [ -1, 6 ]
+  },
+  {
+    name: "post_input2"
+    data_type: TYPE_INT32
+    dims: [ -1 ]
+  }
+]
+
+output [
+  {
+    name: "post_output"
+    data_type: TYPE_STRING
+    dims: [ -1 ]
+  }
+]
+
+instance_group [
+  {
+      count: 1
+      kind: KIND_CPU
+  }
+]

examples/vision/detection/paddledetection/serving/models/postprocess/mask_config.pbtxt

@@ -0,0 +1,34 @@
+backend: "python"
+
+input [
+  {
+    name: "post_input1"
+    data_type: TYPE_FP32
+    dims: [ -1, 6 ]
+  },
+  {
+    name: "post_input2"
+    data_type: TYPE_INT32
+    dims: [ -1 ]
+  },
+  {
+    name: "post_input3"
+    data_type: TYPE_INT32
+    dims: [ -1, -1, -1 ]
+  }
+]
+
+output [
+  {
+    name: "post_output"
+    data_type: TYPE_STRING
+    dims: [ -1 ]
+  }
+]
+
+instance_group [
+  {
+      count: 1
+      kind: KIND_CPU
+  }
+]

examples/vision/detection/paddledetection/serving/models/ppdet/1/README.md

@@ -0,0 +1,3 @@
+# PaddleDetection Pipeline
+
+The pipeline directory does not have model files, but a version number directory needs to be maintained.

examples/vision/detection/paddledetection/serving/models/ppdet/faster_rcnn_config.pbtxt

@@ -0,0 +1,80 @@
+platform: "ensemble"
+
+input [
+  {
+    name: "INPUT"
+    data_type: TYPE_UINT8
+    dims: [ -1, -1, -1, 3 ]
+  }
+]
+output [
+  {
+    name: "DET_RESULT"
+    data_type: TYPE_STRING
+    dims: [ -1 ]
+  }
+]
+ensemble_scheduling {
+  step [
+    {
+      model_name: "preprocess"
+      model_version: 1
+      input_map {
+        key: "preprocess_input"
+        value: "INPUT"
+      }
+      output_map {
+        key: "preprocess_output1"
+        value: "RUNTIME_INPUT1"
+      }
+      output_map {
+        key: "preprocess_output2"
+        value: "RUNTIME_INPUT2"
+      }
+      output_map {
+        key: "preprocess_output3"
+        value: "RUNTIME_INPUT3"
+      }
+    },
+    {
+      model_name: "runtime"
+      model_version: 1
+      input_map {
+        key: "image"
+        value: "RUNTIME_INPUT1"
+      }
+      input_map {
+        key: "scale_factor"
+        value: "RUNTIME_INPUT2"
+      }
+      input_map {
+        key: "im_shape"
+        value: "RUNTIME_INPUT3"
+      }
+      output_map {
+        key: "concat_12.tmp_0"
+        value: "RUNTIME_OUTPUT1"
+      }
+      output_map {
+        key: "concat_8.tmp_0"
+        value: "RUNTIME_OUTPUT2"
+      }
+    },
+    {
+      model_name: "postprocess"
+      model_version: 1
+      input_map {
+        key: "post_input1"
+        value: "RUNTIME_OUTPUT1"
+      }
+      input_map {
+        key: "post_input2"
+        value: "RUNTIME_OUTPUT2"
+      }
+      output_map {
+        key: "post_output"
+        value: "DET_RESULT"
+      }
+    }
+  ]
+}

examples/vision/detection/paddledetection/serving/models/ppdet/mask_rcnn_config.pbtxt

@@ -0,0 +1,88 @@
+platform: "ensemble"
+
+input [
+  {
+    name: "INPUT"
+    data_type: TYPE_UINT8
+    dims: [ -1, -1, -1, 3 ]
+  }
+]
+output [
+  {
+    name: "DET_RESULT"
+    data_type: TYPE_STRING
+    dims: [ -1 ]
+  }
+]
+ensemble_scheduling {
+  step [
+    {
+      model_name: "preprocess"
+      model_version: 1
+      input_map {
+        key: "preprocess_input"
+        value: "INPUT"
+      }
+      output_map {
+        key: "preprocess_output1"
+        value: "RUNTIME_INPUT1"
+      }
+      output_map {
+        key: "preprocess_output2"
+        value: "RUNTIME_INPUT2"
+      }
+      output_map {
+        key: "preprocess_output3"
+        value: "RUNTIME_INPUT3"
+      }
+    },
+    {
+      model_name: "runtime"
+      model_version: 1
+      input_map {
+        key: "image"
+        value: "RUNTIME_INPUT1"
+      }
+      input_map {
+        key: "scale_factor"
+        value: "RUNTIME_INPUT2"
+      }
+      input_map {
+        key: "im_shape"
+        value: "RUNTIME_INPUT3"
+      }
+      output_map {
+        key: "concat_9.tmp_0"
+        value: "RUNTIME_OUTPUT1"
+      }
+      output_map {
+        key: "concat_5.tmp_0"
+        value: "RUNTIME_OUTPUT2"
+      },
+      output_map {
+        key: "tmp_109"
+        value: "RUNTIME_OUTPUT3"
+      }
+    },
+    {
+      model_name: "postprocess"
+      model_version: 1
+      input_map {
+        key: "post_input1"
+        value: "RUNTIME_OUTPUT1"
+      }
+      input_map {
+        key: "post_input2"
+        value: "RUNTIME_OUTPUT2"
+      }
+      input_map {
+        key: "post_input3"
+        value: "RUNTIME_OUTPUT3"
+      }
+      output_map {
+        key: "post_output"
+        value: "DET_RESULT"
+      }
+    }
+  ]
+}