Merge branch 'develop' into optimize-quantization-tool

core/general-client/src/general_model.cpp

@@ -285,22 +285,16 @@ int PredictorClient::batch_predict(
         // int idx = _fetch_name_to_idx[name];
         if (_fetch_name_to_type[name] == 0) {
           VLOG(2) << "ferch var " << name << "type int";
-          model._int64_value_map[name].resize(
-              output.insts(0).tensor_array(idx).int64_data_size());
           int size = output.insts(0).tensor_array(idx).int64_data_size();
-          for (int i = 0; i < size; ++i) {
-            model._int64_value_map[name][i] =
-                output.insts(0).tensor_array(idx).int64_data(i);
-          }
+          model._int64_value_map[name] = std::vector<int64_t>(
+              output.insts(0).tensor_array(idx).int64_data().begin(),
+              output.insts(0).tensor_array(idx).int64_data().begin() + size);
         } else {
           VLOG(2) << "fetch var " << name << "type float";
-          model._float_value_map[name].resize(
-              output.insts(0).tensor_array(idx).float_data_size());
           int size = output.insts(0).tensor_array(idx).float_data_size();
-          for (int i = 0; i < size; ++i) {
-            model._float_value_map[name][i] =
-                output.insts(0).tensor_array(idx).float_data(i);
-          }
+          model._float_value_map[name] = std::vector<float>(
+              output.insts(0).tensor_array(idx).float_data().begin(),
+              output.insts(0).tensor_array(idx).float_data().begin() + size);
         }
         idx += 1;
       }
@@ -564,22 +558,16 @@ int PredictorClient::numpy_predict(
         // int idx = _fetch_name_to_idx[name];
         if (_fetch_name_to_type[name] == 0) {
           VLOG(2) << "ferch var " << name << "type int";
-          model._int64_value_map[name].resize(
-              output.insts(0).tensor_array(idx).int64_data_size());
           int size = output.insts(0).tensor_array(idx).int64_data_size();
-          for (int i = 0; i < size; ++i) {
-            model._int64_value_map[name][i] =
-                output.insts(0).tensor_array(idx).int64_data(i);
-          }
+          model._int64_value_map[name] = std::vector<int64_t>(
+              output.insts(0).tensor_array(idx).int64_data().begin(),
+              output.insts(0).tensor_array(idx).int64_data().begin() + size);
         } else {
           VLOG(2) << "fetch var " << name << "type float";
-          model._float_value_map[name].resize(
-              output.insts(0).tensor_array(idx).float_data_size());
           int size = output.insts(0).tensor_array(idx).float_data_size();
-          for (int i = 0; i < size; ++i) {
-            model._float_value_map[name][i] =
-                output.insts(0).tensor_array(idx).float_data(i);
-          }
+          model._float_value_map[name] = std::vector<float>(
+              output.insts(0).tensor_array(idx).float_data().begin(),
+              output.insts(0).tensor_array(idx).float_data().begin() + size);
         }
         idx += 1;
       }

doc/COMPILE.md

@@ -43,7 +43,7 @@ In the default centos7 image we provide, the Python path is `/usr/bin/python`. I
 ### Integrated CPU version paddle inference library
 
 ``` shell
-mkdir build && cd build
+mkdir server-build-cpu && cd server-build-cpu
 cmake -DPYTHON_INCLUDE_DIR=$PYTHONROOT/include/python2.7/ -DPYTHON_LIBRARIES=$PYTHONROOT/lib/libpython2.7.so -DPYTHON_EXECUTABLE=$PYTHONROOT/bin/python -DSERVER=ON ..
 make -j10
 ```
@@ -53,7 +53,7 @@ you can execute `make install` to put targets under directory `./output`, you ne
 ### Integrated GPU version paddle inference library
 
 ``` shell
-mkdir build && cd build
+mkdir server-build-gpu && cd server-build-gpu
 cmake -DPYTHON_INCLUDE_DIR=$PYTHONROOT/include/python2.7/ -DPYTHON_LIBRARIES=$PYTHONROOT/lib/libpython2.7.so -DPYTHON_EXECUTABLE=$PYTHONROOT/bin/python -DSERVER=ON -DWITH_GPU=ON ..
 make -j10
 ```
@@ -65,7 +65,7 @@ execute `make install` to put targets under directory `./output`
 ## Compile Client
 
 ``` shell
-mkdir build && cd build
+mkdir client-build && cd client-build
 cmake -DPYTHON_INCLUDE_DIR=$PYTHONROOT/include/python2.7/ -DPYTHON_LIBRARIES=$PYTHONROOT/lib/libpython2.7.so -DPYTHON_EXECUTABLE=$PYTHONROOT/bin/python -DCLIENT=ON ..
 make -j10
 ```
@@ -75,7 +75,7 @@ execute `make install` to put targets under directory `./output`
 ## Compile the App
 
 ```bash
-mkdir build && cd build
+mkdir app-build && cd app-build
 cmake -DPYTHON_INCLUDE_DIR=$PYTHONROOT/include/python2.7/ -DPYTHON_LIBRARIES=$PYTHONROOT/lib/libpython2.7.so -DPYTHON_EXECUTABLE=$PYTHONROOT/bin/python -DAPP=ON ..
 make
 ```

doc/COMPILE_CN.md

@@ -43,7 +43,7 @@ export PYTHONROOT=/usr/
 ### 集成CPU版本Paddle Inference Library
 
 ``` shell
-mkdir build && cd build
+mkdir server-build-cpu && cd server-build-cpu
 cmake -DPYTHON_INCLUDE_DIR=$PYTHONROOT/include/python2.7/ -DPYTHON_LIBRARIES=$PYTHONROOT/lib/libpython2.7.so -DPYTHON_EXECUTABLE=$PYTHONROOT/bin/python -DSERVER=ON ..
 make -j10
 ```
@@ -53,7 +53,7 @@ make -j10
 ### 集成GPU版本Paddle Inference Library
 
 ``` shell
-mkdir build && cd build
+mkdir server-build-gpu && cd server-build-gpu
 cmake -DPYTHON_INCLUDE_DIR=$PYTHONROOT/include/python2.7/ -DPYTHON_LIBRARIES=$PYTHONROOT/lib/libpython2.7.so -DPYTHON_EXECUTABLE=$PYTHONROOT/bin/python -DSERVER=ON -DWITH_GPU=ON ..
 make -j10
 ```
@@ -65,7 +65,7 @@ make -j10
 ## 编译Client部分
 
 ``` shell
-mkdir build && cd build
+mkdir client-build && cd client-build
 cmake -DPYTHON_INCLUDE_DIR=$PYTHONROOT/include/python2.7/ -DPYTHON_LIBRARIES=$PYTHONROOT/lib/libpython2.7.so -DPYTHON_EXECUTABLE=$PYTHONROOT/bin/python -DCLIENT=ON ..
 make -j10
 ```
@@ -75,7 +75,7 @@ make -j10
 ## 编译App部分
 
 ```bash
-mkdir build && cd build
+mkdir app-build && cd app-build
 cmake -DPYTHON_INCLUDE_DIR=$PYTHONROOT/include/python2.7/ -DPYTHON_LIBRARIES=$PYTHONROOT/lib/libpython2.7.so -DPYTHON_EXECUTABLE=$PYTHONROOT/bin/python -DCMAKE_INSTALL_PREFIX=./output -DAPP=ON ..
 make
 ```

doc/LATEST_PACKAGES.md

@@ -3,45 +3,45 @@
 ## CPU server
 ### Python 3
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server-0.3.0-py3-none-any.whl
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server-0.3.1-py3-none-any.whl
 ```
 
 ### Python 2
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server-0.3.0-py2-none-any.whl
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server-0.3.1-py2-none-any.whl
 ```
 
 ## GPU server
 ### Python 3
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.0-py3-none-any.whl
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.1-py3-none-any.whl
 ```
 ### Python 2
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.0-py2-none-any.whl
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_server_gpu-0.3.1-py2-none-any.whl
 ```
 
 ## Client
 ### Python 3.7
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_client-0.3.0-cp37-none-manylinux1_x86_64.whl
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_client-0.3.1-cp37-none-any.whl
 ```
 ### Python 3.6
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_client-0.3.0-cp36-none-manylinux1_x86_64.whl
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_client-0.3.1-cp36-none-any.whl
 ```
 ### Python 2.7
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_client-0.3.0-cp27-none-manylinux1_x86_64.whl
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_client-0.3.1-cp27-none-any.whl
 ```
 
 ## App
 ### Python 3
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_app-0.1.0-py3-none-any.whl
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_app-0.1.1-py3-none-any.whl
 ```
 
 ### Python 2
 ```
-https://paddle-serving.bj.bcebos.com/whl/paddle_serving_app-0.1.0-py2-none-any.whl
+https://paddle-serving.bj.bcebos.com/whl/paddle_serving_app-0.1.1-py2-none-any.whl
 ```

python/examples/fit_a_line/test_multilang_client.py

@@ -14,7 +14,10 @@
 # pylint: disable=doc-string-missing
 
 from paddle_serving_client import MultiLangClient
+import functools
 import sys
+import time
+import threading
 
 client = MultiLangClient()
 client.load_client_config(sys.argv[1])
@@ -26,7 +29,22 @@ test_reader = paddle.batch(
         paddle.dataset.uci_housing.test(), buf_size=500),
     batch_size=1)
 
+complete_task_count = [0]
+lock = threading.Lock()
+
+
+def call_back(call_future, data):
+    fetch_map = call_future.result()
+    print("{} {}".format(fetch_map["price"][0], data[0][1][0]))
+    with lock:
+        complete_task_count[0] += 1
+
+
+task_count = 0
 for data in test_reader():
     future = client.predict(feed={"x": data[0][0]}, fetch=["price"], asyn=True)
-    fetch_map = future.result()
-    print("{} {}".format(fetch_map["price"][0], data[0][1][0]))
+    task_count += 1
+    future.add_done_callback(functools.partial(call_back, data=data))
+
+while complete_task_count[0] != task_count:
+    time.sleep(0.1)

python/paddle_serving_app/reader/image_reader.py

@@ -272,8 +272,7 @@ class SegPostprocess(object):
         result_png = score_png
 
         result_png = cv2.resize(
-            result_png,
-            ori_shape[:2],
+            result_png, (ori_shape[1], ori_shape[0]),
             fx=0,
             fy=0,
             interpolation=cv2.INTER_CUBIC)

python/paddle_serving_app/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """ Paddle Serving App version string """
-serving_app_version = "0.1.0"
+serving_app_version = "0.1.1"

python/paddle_serving_client/__init__.py

@@ -391,7 +391,13 @@ class MultiLangClient(object):
         self._parse_model_config(path)
 
     def connect(self, endpoint):
-        self.channel_ = grpc.insecure_channel(endpoint[0])  #TODO
+        # https://github.com/tensorflow/serving/issues/1382
+        options = [('grpc.max_receive_message_length', 512 * 1024 * 1024),
+                   ('grpc.max_send_message_length', 512 * 1024 * 1024),
+                   ('grpc.max_receive_message_length', 512 * 1024 * 1024)]
+
+        self.channel_ = grpc.insecure_channel(
+            endpoint[0], options=options)  #TODO
         self.stub_ = multi_lang_general_model_service_pb2_grpc.MultiLangGeneralModelServiceStub(
             self.channel_)
 
@@ -431,7 +437,6 @@ class MultiLangClient(object):
     def _pack_feed_data(self, feed, fetch, is_python):
         req = multi_lang_general_model_service_pb2.Request()
         req.fetch_var_names.extend(fetch)
-        req.feed_var_names.extend(feed.keys())
         req.is_python = is_python
         feed_batch = None
         if isinstance(feed, dict):
@@ -440,6 +445,7 @@ class MultiLangClient(object):
             feed_batch = feed
         else:
             raise Exception("{} not support".format(type(feed)))
+        req.feed_var_names.extend(feed_batch[0].keys())
         init_feed_names = False
         for feed_data in feed_batch:
             inst = multi_lang_general_model_service_pb2.FeedInst()
@@ -516,6 +522,9 @@ class MultiLangClient(object):
 
         return unpack_resp
 
+    def get_feed_names(self):
+        return self.feed_names_
+
     def predict(self,
                 feed,
                 fetch,
@@ -548,3 +557,10 @@ class MultiLangPredictFuture(object):
     def result(self):
         resp = self.call_future_.result()
         return self.callback_func_(resp)
+
+    def add_done_callback(self, fn):
+        def __fn__(call_future):
+            assert call_future == self.call_future_
+            fn(self)
+
+        self.call_future_.add_done_callback(__fn__)

python/paddle_serving_client/version.py

@@ -12,6 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """ Paddle Serving Client version string """
-serving_client_version = "0.3.0"
-serving_server_version = "0.3.0"
-module_proto_version = "0.3.0"
+serving_client_version = "0.3.1"
+serving_server_version = "0.3.1"
+module_proto_version = "0.3.1"

python/paddle_serving_server/version.py

@@ -12,6 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """ Paddle Serving Client version string """
-serving_client_version = "0.3.0"
-serving_server_version = "0.3.0"
-module_proto_version = "0.3.0"
+serving_client_version = "0.3.1"
+serving_server_version = "0.3.1"
+module_proto_version = "0.3.1"

python/paddle_serving_server_gpu/version.py

@@ -12,6 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """ Paddle Serving Client version string """
-serving_client_version = "0.3.0"
-serving_server_version = "0.3.0"
-module_proto_version = "0.3.0"
+serving_client_version = "0.3.1"
+serving_server_version = "0.3.1"
+module_proto_version = "0.3.1"

tools/python_tag.py

@@ -15,6 +15,6 @@
 from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag
 import re
 with open("setup.cfg", "w") as f:
-    line = "[bdist_wheel]\npython-tag={0}{1}\nplat-name=manylinux1_x86_64".format(
-        get_abbr_impl(), get_impl_ver())
+    line = "[bdist_wheel]\npython-tag={0}{1}".format(get_abbr_impl(),
+                                                     get_impl_ver())
     f.write(line)