change the unit of auto_batching_timeout to ms

doc/PIPELINE_SERVING.md

@@ -98,7 +98,7 @@ The meaning of each parameter is as follows:
 |        timeout        | (int) The timeout time of the process operation, in seconds. 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.    |
+| auto_batching_timeout | (float) Timeout for building batches of Auto-Batching (the unit is ms). |
 
 
 #### 2. General OP Secondary Development Interface

doc/PIPELINE_SERVING_CN.md

@@ -98,7 +98,7 @@ def __init__(name=None,
 |        timeout        | （int）process 操作的超时时间，单位为秒。若该值小于零，则视作不超时。 |
 |         retry         |       （int）超时重试次数。当该值为 1 时，不进行重试。       |
 |      batch_size       | （int）进行 Auto-Batching 的期望 batch_size 大小，由于构建 batch 可能超时，实际 batch_size 可能小于设定值。 |
-| auto_batching_timeout |     （float）进行 Auto-Batching 构建 batch 的超时时间。      |
+| auto_batching_timeout | （float）进行 Auto-Batching 构建 batch 的超时时间，单位为毫秒。 |
 
 
 #### 2. 普通 OP二次开发接口

python/examples/pipeline/imdb_model_ensemble/test_pipeline_client.py

@@ -11,7 +11,7 @@
 # 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.
-from paddle_serving_server_gpu.pipeline import PipelineClient
+from paddle_serving_server.pipeline import PipelineClient
 import numpy as np
 
 client = PipelineClient()

python/examples/pipeline/imdb_model_ensemble/test_pipeline_server.py

@@ -15,13 +15,13 @@
 
 import logging
 logging.basicConfig(
-    format="[%(process)d](%(threadName)s) %(levelname)s %(asctime)s [%(funcName)s:%(lineno)d] %(message)s",
+    format="[%(process)d](%(threadName)s) %(levelname)s %(asctime)s [%(filename)s:%(lineno)d] %(message)s",
     level=logging.INFO)
 
-from paddle_serving_server_gpu.pipeline import Op, RequestOp, ResponseOp
-from paddle_serving_server_gpu.pipeline import PipelineServer
-from paddle_serving_server_gpu.pipeline.proto import pipeline_service_pb2
-from paddle_serving_server_gpu.pipeline.channel import ChannelDataEcode
+from paddle_serving_server.pipeline import Op, RequestOp, ResponseOp
+from paddle_serving_server.pipeline import PipelineServer
+from paddle_serving_server.pipeline.proto import pipeline_service_pb2
+from paddle_serving_server.pipeline.channel import ChannelDataEcode
 import numpy as np
 from paddle_serving_app.reader import IMDBDataset
 
@@ -80,7 +80,7 @@ bow_op = Op(name="bow",
             timeout=-1,
             retry=1,
             batch_size=3,
-            auto_batching_timeout=1)
+            auto_batching_timeout=1000)
 cnn_op = Op(name="cnn",
             input_ops=[read_op],
             server_endpoints=["127.0.0.1:9292"],

python/pipeline/operator.py

@@ -70,6 +70,8 @@ class Op(object):
         if self._auto_batching_timeout is not None:
             if self._auto_batching_timeout <= 0 or self._batch_size == 1:
                 self._auto_batching_timeout = None
+            else:
+                self._auto_batching_timeout = self._auto_batching_timeout / 1000.0
 
         self._server_use_profile = False