Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-prefetch

.clang-format

@@ -19,7 +19,7 @@ BasedOnStyle:  Google
 IndentWidth:     2
 TabWidth:        2
 ContinuationIndentWidth: 4
-AccessModifierOffset: -2  # The private/protected/public has no indent in class
+AccessModifierOffset: -1  # The private/protected/public has no indent in class
 Standard:  Cpp11 
 AllowAllParametersOfDeclarationOnNextLine: true
 BinPackParameters: false

.pre-commit-config.yaml

@@ -34,6 +34,14 @@ repos:
         entry: bash ./tools/codestyle/cpplint_pre_commit.hook
         language: system
         files: \.(c|cc|cxx|cpp|cu|h|hpp|hxx)$
+-   repo: local
+    hooks:
+    -   id: pylint-doc-string
+        name: pylint
+        description: Check python docstring style using docstring_checker.
+        entry: bash ./tools/codestyle/pylint_pre_commit.hook
+        language: system
+        files: \.(py)$
 -   repo: https://github.com/PaddlePaddle/pre-commit-golang
     sha: 8337620115c25ff8333f1b1a493bd031049bd7c0
     hooks:

.travis.yml

@@ -18,6 +18,8 @@ env:
 addons:
   ssh_known_hosts: 13.229.163.131
 before_install:
+  # For pylint dockstring checker
+  - sudo pip install pylint pytest astroid isort
   - |
     function timeout() { perl -e 'alarm shift; exec @ARGV' "$@"; }
 script:

Dockerfile

@@ -79,6 +79,9 @@ RUN pip install pre-commit 'ipython==5.3.0' && \
     pip install 'ipykernel==4.6.0' 'jupyter==1.0.0' && \
     pip install opencv-python
 
+#For docstring checker
+RUN pip install pylint pytest astroid isort
+
 COPY ./python/requirements.txt /root/
 RUN pip install -r /root/requirements.txt
 

benchmark/fluid/README.md

@@ -24,22 +24,22 @@ Currently supported `--model` argument include:
 
 * Run the following command to start a benchmark job locally:
     ```bash
-      python fluid_benchmark.py --model mnist --parallel 1 --device GPU --with_test
+      python fluid_benchmark.py --model mnist  --device GPU
     ```
     You can choose to use GPU/CPU training. With GPU training, you can specify
-    `--parallel 1` to run multi GPU training.
+    `--gpus <gpu_num>` to run multi GPU training.
 * Run distributed training with parameter servers:
     * start parameter servers:
         ```bash
-        PADDLE_TRAINING_ROLE=PSERVER PADDLE_PSERVER_PORT=7164 PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=1 PADDLE_CURRENT_IP=127.0.0.1 PADDLE_TRAINER_ID=0 python fluid_benchmark.py --model mnist --parallel 0 --device GPU --update_method pserver
+        PADDLE_TRAINING_ROLE=PSERVER PADDLE_PSERVER_PORT=7164 PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=1 PADDLE_CURRENT_IP=127.0.0.1 PADDLE_TRAINER_ID=0 python fluid_benchmark.py --model mnist  --device GPU --update_method pserver
         ```
     * start trainers:
         ```bash
-        PADDLE_TRAINING_ROLE=PSERVER PADDLE_PSERVER_PORT=7164 PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=1 PADDLE_CURRENT_IP=127.0.0.1 PADDLE_TRAINER_ID=0 python fluid_benchmark.py --model mnist --parallel 0 --device GPU --update_method pserver
+        PADDLE_TRAINING_ROLE=TRAINER PADDLE_PSERVER_PORT=7164 PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=1 PADDLE_CURRENT_IP=127.0.0.1 PADDLE_TRAINER_ID=0 python fluid_benchmark.py --model mnist  --device GPU --update_method pserver
         ```
 * Run distributed training using NCCL2
     ```bash
-    PADDLE_PSERVER_PORT=7164 PADDLE_TRAINER_IPS=192.168.0.2,192.168.0.3  PADDLE_CURRENT_IP=127.0.0.1 PADDLE_TRAINER_ID=0 python fluid_benchmark.py --model mnist --parallel 0 --device GPU --update_method nccl2
+    PADDLE_PSERVER_PORT=7164 PADDLE_TRAINER_IPS=192.168.0.2,192.168.0.3  PADDLE_CURRENT_IP=127.0.0.1 PADDLE_TRAINER_ID=0 python fluid_benchmark.py --model mnist --device GPU --update_method nccl2
     ```
 
 ## Run Distributed Benchmark on Kubernetes Cluster
@@ -48,7 +48,7 @@ We provide a script `kube_gen_job.py` to generate Kubernetes yaml files to submi
 distributed benchmark jobs to your cluster. To generate a job yaml, just run:
 
 ```bash
-python kube_gen_job.py --jobname myjob --pscpu 4 --cpu 8 --gpu 8 --psmemory 20 --memory 40 --pservers 4 --trainers 4 --entry "python fluid_benchmark.py --model mnist --parallel 1 --device GPU --update_method pserver --with_test" --disttype pserver
+python kube_gen_job.py --jobname myjob --pscpu 4 --cpu 8 --gpu 8 --psmemory 20 --memory 40 --pservers 4 --trainers 4 --entry "python fluid_benchmark.py --model mnist --parallel 1 --device GPU --update_method pserver " --disttype pserver
 ```
 
 Then the yaml files are generated under directory `myjob`, you can run:

cmake/inference_lib.cmake

@@ -56,24 +56,28 @@ set(dst_dir "${FLUID_INSTALL_DIR}/third_party/eigen3")
 copy(eigen3_lib
   SRCS ${EIGEN_INCLUDE_DIR}/Eigen/Core ${EIGEN_INCLUDE_DIR}/Eigen/src ${EIGEN_INCLUDE_DIR}/unsupported/Eigen
   DSTS ${dst_dir}/Eigen ${dst_dir}/Eigen ${dst_dir}/unsupported
+  DEPS eigen3
 )
 
 set(dst_dir "${FLUID_INSTALL_DIR}/third_party/install/gflags")
 copy(gflags_lib
   SRCS ${GFLAGS_INCLUDE_DIR} ${GFLAGS_LIBRARIES}
   DSTS ${dst_dir} ${dst_dir}/lib
+  DEPS gflags
 )
 
 set(dst_dir "${FLUID_INSTALL_DIR}/third_party/install/glog")
 copy(glog_lib
   SRCS ${GLOG_INCLUDE_DIR} ${GLOG_LIBRARIES}
   DSTS ${dst_dir} ${dst_dir}/lib
+  DEPS glog
 )
 
 set(dst_dir "${FLUID_INSTALL_DIR}/third_party/boost/")
 copy(boost_lib
   SRCS ${BOOST_INCLUDE_DIR}/boost
   DSTS ${dst_dir}
+  DEPS boost
 )
 
 if(NOT PROTOBUF_FOUND)
@@ -81,6 +85,7 @@ if(NOT PROTOBUF_FOUND)
     copy(protobuf_lib
       SRCS ${PROTOBUF_INCLUDE_DIR} ${PROTOBUF_LIBRARY}
       DSTS ${dst_dir} ${dst_dir}/lib
+      DEPS extern_protobuf
     )
 endif()
 
@@ -89,12 +94,14 @@ if(NOT CBLAS_FOUND)
     copy(openblas_lib
       SRCS ${CBLAS_INSTALL_DIR}/lib ${CBLAS_INSTALL_DIR}/include
       DSTS ${dst_dir} ${dst_dir}
+      DEPS extern_openblas
     )
 elseif (WITH_MKLML)
     set(dst_dir "${FLUID_INSTALL_DIR}/third_party/install/mklml")
     copy(mklml_lib
       SRCS ${MKLML_LIB} ${MKLML_IOMP_LIB} ${MKLML_INC_DIR}
       DSTS ${dst_dir}/lib ${dst_dir}/lib ${dst_dir}
+      DEPS mklml
     )
 endif()
 
@@ -103,6 +110,7 @@ if(WITH_MKLDNN)
   copy(mkldnn_lib
     SRCS ${MKLDNN_INC_DIR} ${MKLDNN_SHARED_LIB}
     DSTS ${dst_dir} ${dst_dir}/lib
+    DEPS mkldnn
   )
 endif()
 
@@ -110,17 +118,20 @@ if(NOT MOBILE_INFERENCE AND NOT RPI)
   set(dst_dir "${FLUID_INSTALL_DIR}/third_party/install/snappy")
   copy(snappy_lib
     SRCS ${SNAPPY_INCLUDE_DIR} ${SNAPPY_LIBRARIES}
-    DSTS ${dst_dir} ${dst_dir}/lib)
+    DSTS ${dst_dir} ${dst_dir}/lib
+    DEPS snappy)
 
   set(dst_dir "${FLUID_INSTALL_DIR}/third_party/install/snappystream")
   copy(snappystream_lib
     SRCS ${SNAPPYSTREAM_INCLUDE_DIR} ${SNAPPYSTREAM_LIBRARIES}
-    DSTS ${dst_dir} ${dst_dir}/lib)
+    DSTS ${dst_dir} ${dst_dir}/lib
+    DEPS snappystream)
 
   set(dst_dir "${FLUID_INSTALL_DIR}/third_party/install/zlib")
   copy(zlib_lib
     SRCS ${ZLIB_INCLUDE_DIR} ${ZLIB_LIBRARIES}
-    DSTS ${dst_dir} ${dst_dir}/lib)
+    DSTS ${dst_dir} ${dst_dir}/lib
+    DEPS zlib)
 endif()
 
 # paddle fluid module