Merge remote-tracking branch 'origin/dygraph' into dygraph

deploy/cpp_infer/CMakeLists.txt

@@ -206,6 +206,10 @@ endif()
 
 set(DEPS ${DEPS} ${OpenCV_LIBS})
 
+include(ExternalProject)
+include(external-cmake/auto-log.cmake)
+include_directories(${CMAKE_CURRENT_BINARY_DIR}/autolog/src/extern_Autolog/auto_log)
+
 AUX_SOURCE_DIRECTORY(./src SRCS)
 add_executable(${DEMO_NAME} ${SRCS})
 

deploy/cpp_infer/external-cmake/auto-log.cmake

+find_package(Git REQUIRED)
+message("${CMAKE_BUILD_TYPE}")
+
+set(AUTOLOG_REPOSITORY     https://github.com/LDOUBLEV/AutoLog.git)
+SET(AUTOLOG_INSTALL_DIR   ${CMAKE_CURRENT_BINARY_DIR}/install/Autolog)
+
+ExternalProject_Add(
+    extern_Autolog
+    PREFIX autolog
+    GIT_REPOSITORY ${AUTOLOG_REPOSITORY}
+    GIT_TAG main
+    DOWNLOAD_NO_EXTRACT True
+    INSTALL_COMMAND cmake -E echo "Skipping install step."
+)

deploy/cpp_infer/src/main.cpp

@@ -39,8 +39,8 @@
 DEFINE_bool(use_gpu, false, "Infering with GPU or CPU.");
 DEFINE_int32(gpu_id, 0, "Device id of GPU to execute.");
 DEFINE_int32(gpu_mem, 4000, "GPU id when infering with GPU.");
-DEFINE_int32(cpu_math_library_num_threads, 10, "Num of threads with CPU.");
-DEFINE_bool(use_mkldnn, false, "Whether use mkldnn with CPU.");
+DEFINE_int32(cpu_threads, 10, "Num of threads with CPU.");
+DEFINE_bool(enable_mkldnn, false, "Whether use mkldnn with CPU.");
 DEFINE_bool(use_tensorrt, false, "Whether use tensorrt.");
 DEFINE_string(precision, "fp32", "Precision be one of fp32/fp16/int8");
 DEFINE_bool(benchmark, true, "Whether use benchmark.");
@@ -60,6 +60,7 @@ DEFINE_string(cls_model_dir, "", "Path of cls inference model.");
 DEFINE_double(cls_thresh, 0.9, "Threshold of cls_thresh.");
 // recognition related
 DEFINE_string(rec_model_dir, "", "Path of rec inference model.");
+DEFINE_int32(rec_batch_num, 1, "rec_batch_num.");
 DEFINE_string(char_list_file, "../../ppocr/utils/ppocr_keys_v1.txt", "Path of dictionary.");
 
 
@@ -68,34 +69,6 @@ using namespace cv;
 using namespace PaddleOCR;
 
 
-void PrintBenchmarkLog(std::string model_name, 
-                       int batch_size, 
-                       std::string input_shape,
-                       std::vector<double> time_info,
-                       int img_num){
-  LOG(INFO) << "----------------------- Config info -----------------------";
-  LOG(INFO) << "runtime_device: " << (FLAGS_use_gpu ? "gpu" : "cpu");
-  LOG(INFO) << "ir_optim: " << "True";
-  LOG(INFO) << "enable_memory_optim: " << "True";
-  LOG(INFO) << "enable_tensorrt: " << FLAGS_use_tensorrt;
-  LOG(INFO) << "enable_mkldnn: " << (FLAGS_use_mkldnn ? "True" : "False");
-  LOG(INFO) << "cpu_math_library_num_threads: " << FLAGS_cpu_math_library_num_threads;
-  LOG(INFO) << "----------------------- Data info -----------------------";
-  LOG(INFO) << "batch_size: " << batch_size;
-  LOG(INFO) << "input_shape: " << input_shape;
-  LOG(INFO) << "data_num: " << img_num;
-  LOG(INFO) << "----------------------- Model info -----------------------";
-  LOG(INFO) << "model_name: " << model_name;
-  LOG(INFO) << "precision: " << FLAGS_precision;
-  LOG(INFO) << "----------------------- Perf info ------------------------";
-  LOG(INFO) << "Total time spent(ms): "
-            << std::accumulate(time_info.begin(), time_info.end(), 0);
-  LOG(INFO) << "preprocess_time(ms): " << time_info[0] / img_num
-            << ", inference_time(ms): " << time_info[1] / img_num
-            << ", postprocess_time(ms): " << time_info[2] / img_num;
-}
-
-
 static bool PathExists(const std::string& path){
 #ifdef _WIN32
   struct _stat buffer;
@@ -110,8 +83,8 @@ static bool PathExists(const std::string& path){
 int main_det(std::vector<cv::String> cv_all_img_names) {
     std::vector<double> time_info = {0, 0, 0};
     DBDetector det(FLAGS_det_model_dir, FLAGS_use_gpu, FLAGS_gpu_id,
-                   FLAGS_gpu_mem, FLAGS_cpu_math_library_num_threads, 
-                   FLAGS_use_mkldnn, FLAGS_max_side_len, FLAGS_det_db_thresh,
+                   FLAGS_gpu_mem, FLAGS_cpu_threads, 
+                   FLAGS_enable_mkldnn, FLAGS_max_side_len, FLAGS_det_db_thresh,
                    FLAGS_det_db_box_thresh, FLAGS_det_db_unclip_ratio,
                    FLAGS_use_polygon_score, FLAGS_visualize,
                    FLAGS_use_tensorrt, FLAGS_precision);
@@ -135,7 +108,17 @@ int main_det(std::vector<cv::String> cv_all_img_names) {
     }
     
     if (FLAGS_benchmark) {
-        PrintBenchmarkLog("det", 1, "dynamic", time_info, cv_all_img_names.size());
+        AutoLogger autolog("ocr_det", 
+                           FLAGS_use_gpu,
+                           FLAGS_use_tensorrt,
+                           FLAGS_enable_mkldnn,
+                           FLAGS_cpu_threads,
+                           1, 
+                           "dynamic", 
+                           FLAGS_precision, 
+                           time_info, 
+                           cv_all_img_names.size());
+        autolog.report();
     }
     return 0;
 }
@@ -144,8 +127,8 @@ int main_det(std::vector<cv::String> cv_all_img_names) {
 int main_rec(std::vector<cv::String> cv_all_img_names) {
     std::vector<double> time_info = {0, 0, 0};
     CRNNRecognizer rec(FLAGS_rec_model_dir, FLAGS_use_gpu, FLAGS_gpu_id,
-                       FLAGS_gpu_mem, FLAGS_cpu_math_library_num_threads,
-                       FLAGS_use_mkldnn, FLAGS_char_list_file,
+                       FLAGS_gpu_mem, FLAGS_cpu_threads,
+                       FLAGS_enable_mkldnn, FLAGS_char_list_file,
                        FLAGS_use_tensorrt, FLAGS_precision);
 
     for (int i = 0; i < cv_all_img_names.size(); ++i) {
@@ -165,18 +148,14 @@ int main_rec(std::vector<cv::String> cv_all_img_names) {
       time_info[2] += rec_times[2];
     }
     
-    if (FLAGS_benchmark) {
-        PrintBenchmarkLog("rec", 1, "dynamic", time_info, cv_all_img_names.size());
-    }
-    
     return 0;
 }
 
 
 int main_system(std::vector<cv::String> cv_all_img_names) {
     DBDetector det(FLAGS_det_model_dir, FLAGS_use_gpu, FLAGS_gpu_id,
-                   FLAGS_gpu_mem, FLAGS_cpu_math_library_num_threads, 
-                   FLAGS_use_mkldnn, FLAGS_max_side_len, FLAGS_det_db_thresh,
+                   FLAGS_gpu_mem, FLAGS_cpu_threads, 
+                   FLAGS_enable_mkldnn, FLAGS_max_side_len, FLAGS_det_db_thresh,
                    FLAGS_det_db_box_thresh, FLAGS_det_db_unclip_ratio,
                    FLAGS_use_polygon_score, FLAGS_visualize,
                    FLAGS_use_tensorrt, FLAGS_precision);
@@ -184,14 +163,14 @@ int main_system(std::vector<cv::String> cv_all_img_names) {
     Classifier *cls = nullptr;
     if (FLAGS_use_angle_cls) {
       cls = new Classifier(FLAGS_cls_model_dir, FLAGS_use_gpu, FLAGS_gpu_id,
-                           FLAGS_gpu_mem, FLAGS_cpu_math_library_num_threads,
-                           FLAGS_use_mkldnn, FLAGS_cls_thresh,
+                           FLAGS_gpu_mem, FLAGS_cpu_threads,
+                           FLAGS_enable_mkldnn, FLAGS_cls_thresh,
                            FLAGS_use_tensorrt, FLAGS_precision);
     }
 
     CRNNRecognizer rec(FLAGS_rec_model_dir, FLAGS_use_gpu, FLAGS_gpu_id,
-                       FLAGS_gpu_mem, FLAGS_cpu_math_library_num_threads,
-                       FLAGS_use_mkldnn, FLAGS_char_list_file,
+                       FLAGS_gpu_mem, FLAGS_cpu_threads,
+                       FLAGS_enable_mkldnn, FLAGS_char_list_file,
                        FLAGS_use_tensorrt, FLAGS_precision);
 
     auto start = std::chrono::system_clock::now();

ppocr/losses/cls_loss.py

@@ -25,6 +25,6 @@ class ClsLoss(nn.Layer):
         self.loss_func = nn.CrossEntropyLoss(reduction='mean')
 
     def forward(self, predicts, batch):
-        label = batch[1]
+        label = batch[1].astype("int64")
         loss = self.loss_func(input=predicts, label=label)
         return {'loss': loss}

tests/ocr_det_params.txt

@@ -49,4 +49,19 @@ inference:tools/infer/predict_det.py
 --save_log_path:null
 --benchmark:True
 null:null
+===========================cpp_infer_params===========================
+use_opencv:True
+infer_model:./inference/ch_ppocr_mobile_v2.0_det_infer/
+infer_quant:False
+inference:./deploy/cpp_infer/build/ppocr det
+--use_gpu:True|False
+--enable_mkldnn:True|False
+--cpu_threads:1|6
+--rec_batch_num:1
+--use_tensorrt:False|True
+--precision:fp32|fp16
+--det_model_dir:
+--image_dir:./inference/ch_det_data_50/all-sum-510/
+--save_log_path:null
+--benchmark:True
 

tests/prepare.sh

 #!/bin/bash
 FILENAME=$1
-# MODE be one of ['lite_train_infer' 'whole_infer' 'whole_train_infer', 'infer']
+# MODE be one of ['lite_train_infer' 'whole_infer' 'whole_train_infer', 'infer', 'cpp_infer']
 MODE=$2
 
 dataline=$(cat ${FILENAME})
@@ -58,7 +58,7 @@ elif [ ${MODE} = "whole_infer" ];then
     cd ./train_data/ && tar xf icdar2015_infer.tar && tar xf ic15_data.tar
     ln -s ./icdar2015_infer ./icdar2015
     cd ../
-else
+elif [ ${MODE} = "infer" ] || [ ${MODE} = "cpp_infer" ];then
     if [ ${model_name} = "ocr_det" ]; then
         eval_model_name="ch_ppocr_mobile_v2.0_det_infer"
         rm -rf ./train_data/icdar2015
@@ -74,3 +74,72 @@ else
     fi 
 fi
 
+if [ ${MODE} = "cpp_infer" ];then
+    cd deploy/cpp_infer
+    use_opencv=$(func_parser_value "${lines[52]}")
+    if [ ${use_opencv} = "True" ]; then
+        echo "################### build opencv ###################"
+        rm -rf 3.4.7.tar.gz opencv-3.4.7/
+        wget https://github.com/opencv/opencv/archive/3.4.7.tar.gz
+        tar -xf 3.4.7.tar.gz
+
+        cd opencv-3.4.7/
+        install_path=$(pwd)/opencv-3.4.7/opencv3
+
+        rm -rf build
+        mkdir build
+        cd build
+
+        cmake .. \
+            -DCMAKE_INSTALL_PREFIX=${install_path} \
+            -DCMAKE_BUILD_TYPE=Release \
+            -DBUILD_SHARED_LIBS=OFF \
+            -DWITH_IPP=OFF \
+            -DBUILD_IPP_IW=OFF \
+            -DWITH_LAPACK=OFF \
+            -DWITH_EIGEN=OFF \
+            -DCMAKE_INSTALL_LIBDIR=lib64 \
+            -DWITH_ZLIB=ON \
+            -DBUILD_ZLIB=ON \
+            -DWITH_JPEG=ON \
+            -DBUILD_JPEG=ON \
+            -DWITH_PNG=ON \
+            -DBUILD_PNG=ON \
+            -DWITH_TIFF=ON \
+            -DBUILD_TIFF=ON
+
+        make -j
+        make install
+        cd ../
+        echo "################### build opencv finished ###################"
+    fi
+
+
+    echo "################### build PaddleOCR demo ####################"
+    if [ ${use_opencv} = "True" ]; then
+        OPENCV_DIR=$(pwd)/opencv-3.4.7/opencv3/
+    else
+        OPENCV_DIR=''
+    fi
+    LIB_DIR=$(pwd)/Paddle/build/paddle_inference_install_dir/
+    CUDA_LIB_DIR=$(dirname `find /usr -name libcudart.so`)
+    CUDNN_LIB_DIR=$(dirname `find /usr -name libcudnn.so`)
+    
+    BUILD_DIR=build
+    rm -rf ${BUILD_DIR}
+    mkdir ${BUILD_DIR}
+    cd ${BUILD_DIR}
+    cmake .. \
+        -DPADDLE_LIB=${LIB_DIR} \
+        -DWITH_MKL=ON \
+        -DWITH_GPU=OFF \
+        -DWITH_STATIC_LIB=OFF \
+        -DWITH_TENSORRT=OFF \
+        -DOPENCV_DIR=${OPENCV_DIR} \
+        -DCUDNN_LIB=${CUDNN_LIB_DIR} \
+        -DCUDA_LIB=${CUDA_LIB_DIR} \
+        -DTENSORRT_DIR=${TENSORRT_DIR} \
+
+    make -j
+    echo "################### build PaddleOCR demo finished ###################"
+fi
\ No newline at end of file

tests/test.sh

 #!/bin/bash
 FILENAME=$1
-# MODE be one of ['lite_train_infer' 'whole_infer' 'whole_train_infer', 'infer']
+# MODE be one of ['lite_train_infer' 'whole_infer' 'whole_train_infer', 'infer', 'cpp_infer']
 MODE=$2
 
 dataline=$(cat ${FILENAME})
@@ -145,6 +145,33 @@ benchmark_value=$(func_parser_value "${lines[49]}")
 infer_key1=$(func_parser_key "${lines[50]}")
 infer_value1=$(func_parser_value "${lines[50]}")
 
+if [ ${MODE} = "cpp_infer" ]; then
+    # parser cpp inference model 
+    cpp_infer_model_dir_list=$(func_parser_value "${lines[53]}")
+    cpp_infer_is_quant=$(func_parser_value "${lines[54]}")
+    # parser cpp inference 
+    inference_cmd=$(func_parser_value "${lines[55]}")
+    cpp_use_gpu_key=$(func_parser_key "${lines[56]}")
+    cpp_use_gpu_list=$(func_parser_value "${lines[56]}")
+    cpp_use_mkldnn_key=$(func_parser_key "${lines[57]}")
+    cpp_use_mkldnn_list=$(func_parser_value "${lines[57]}")
+    cpp_cpu_threads_key=$(func_parser_key "${lines[58]}")
+    cpp_cpu_threads_list=$(func_parser_value "${lines[58]}")
+    cpp_batch_size_key=$(func_parser_key "${lines[59]}")
+    cpp_batch_size_list=$(func_parser_value "${lines[59]}")
+    cpp_use_trt_key=$(func_parser_key "${lines[60]}")
+    cpp_use_trt_list=$(func_parser_value "${lines[60]}")
+    cpp_precision_key=$(func_parser_key "${lines[61]}")
+    cpp_precision_list=$(func_parser_value "${lines[61]}")
+    cpp_infer_model_key=$(func_parser_key "${lines[62]}")
+    cpp_image_dir_key=$(func_parser_key "${lines[63]}")
+    cpp_infer_img_dir=$(func_parser_value "${lines[63]}")
+    cpp_save_log_key=$(func_parser_key "${lines[64]}")
+    cpp_benchmark_key=$(func_parser_key "${lines[65]}")
+    cpp_benchmark_value=$(func_parser_value "${lines[65]}")
+fi
+
+
 LOG_PATH="./tests/output"
 mkdir -p ${LOG_PATH}
 status_log="${LOG_PATH}/results.log"
@@ -218,6 +245,71 @@ function func_inference(){
     done
 }
 
+function func_cpp_inference(){
+    IFS='|'
+    _script=$1
+    _model_dir=$2
+    _log_path=$3
+    _img_dir=$4
+    _flag_quant=$5
+    # inference 
+    for use_gpu in ${cpp_use_gpu_list[*]}; do
+        if [ ${use_gpu} = "False" ] || [ ${use_gpu} = "cpu" ]; then
+            for use_mkldnn in ${cpp_use_mkldnn_list[*]}; do
+                if [ ${use_mkldnn} = "False" ] && [ ${_flag_quant} = "True" ]; then
+                    continue
+                fi
+                for threads in ${cpp_cpu_threads_list[*]}; do
+                    for batch_size in ${cpp_batch_size_list[*]}; do
+                        _save_log_path="${_log_path}/cpp_infer_cpu_usemkldnn_${use_mkldnn}_threads_${threads}_batchsize_${batch_size}.log"
+                        set_infer_data=$(func_set_params "${cpp_image_dir_key}" "${_img_dir}")
+                        set_benchmark=$(func_set_params "${cpp_benchmark_key}" "${cpp_benchmark_value}")
+                        set_batchsize=$(func_set_params "${cpp_batch_size_key}" "${batch_size}")
+                        set_cpu_threads=$(func_set_params "${cpp_cpu_threads_key}" "${threads}")
+                        set_model_dir=$(func_set_params "${cpp_infer_model_key}" "${_model_dir}")
+                        command="${_script} ${cpp_use_gpu_key}=${use_gpu} ${cpp_use_mkldnn_key}=${use_mkldnn} ${set_cpu_threads} ${set_model_dir} ${set_batchsize} ${set_infer_data} ${set_benchmark} > ${_save_log_path} 2>&1 "
+                        eval $command
+                        last_status=${PIPESTATUS[0]}
+                        eval "cat ${_save_log_path}"
+                        status_check $last_status "${command}" "${status_log}"
+                    done
+                done
+            done
+        elif [ ${use_gpu} = "True" ] || [ ${use_gpu} = "gpu" ]; then
+            for use_trt in ${cpp_use_trt_list[*]}; do
+                for precision in ${cpp_precision_list[*]}; do
+                    if [[ ${_flag_quant} = "False" ]] && [[ ${precision} =~ "int8" ]]; then
+                        continue
+                    fi 
+                    if [[ ${precision} =~ "fp16" || ${precision} =~ "int8" ]] && [ ${use_trt} = "False" ]; then
+                        continue
+                    fi
+                    if [[ ${use_trt} = "False" || ${precision} =~ "int8" ]] && [ ${_flag_quant} = "True" ]; then
+                        continue
+                    fi
+                    for batch_size in ${cpp_batch_size_list[*]}; do
+                        _save_log_path="${_log_path}/cpp_infer_gpu_usetrt_${use_trt}_precision_${precision}_batchsize_${batch_size}.log"
+                        set_infer_data=$(func_set_params "${cpp_image_dir_key}" "${_img_dir}")
+                        set_benchmark=$(func_set_params "${cpp_benchmark_key}" "${cpp_benchmark_value}")
+                        set_batchsize=$(func_set_params "${cpp_batch_size_key}" "${batch_size}")
+                        set_tensorrt=$(func_set_params "${cpp_use_trt_key}" "${use_trt}")
+                        set_precision=$(func_set_params "${cpp_precision_key}" "${precision}")
+                        set_model_dir=$(func_set_params "${cpp_infer_model_key}" "${_model_dir}")
+                        command="${_script} ${cpp_use_gpu_key}=${use_gpu} ${set_tensorrt} ${set_precision} ${set_model_dir} ${set_batchsize} ${set_infer_data} ${set_benchmark} > ${_save_log_path} 2>&1 "
+                        eval $command
+                        last_status=${PIPESTATUS[0]}
+                        eval "cat ${_save_log_path}"
+                        status_check $last_status "${command}" "${status_log}"
+                        
+                    done
+                done
+            done
+        else
+            echo "Does not support hardware other than CPU and GPU Currently!"
+        fi
+    done
+}
+
 if [ ${MODE} = "infer" ]; then
     GPUID=$3
     if [ ${#GPUID} -le 0 ];then
@@ -252,6 +344,25 @@ if [ ${MODE} = "infer" ]; then
         Count=$(($Count + 1))
     done
 
+elif [ ${MODE} = "cpp_infer" ]; then
+    GPUID=$3
+    if [ ${#GPUID} -le 0 ];then
+        env=" "
+    else
+        env="export CUDA_VISIBLE_DEVICES=${GPUID}"
+    fi
+    # set CUDA_VISIBLE_DEVICES
+    eval $env
+    export Count=0
+    IFS="|"
+    infer_quant_flag=(${cpp_infer_is_quant})
+    for infer_model in ${cpp_infer_model_dir_list[*]}; do
+        #run inference
+        is_quant=${infer_quant_flag[Count]}
+        func_cpp_inference "${inference_cmd}" "${infer_model}" "${LOG_PATH}" "${cpp_infer_img_dir}" ${is_quant}
+        Count=$(($Count + 1))
+    done
+
 else
     IFS="|"
     export Count=0

tools/infer/predict_rec.py

@@ -278,7 +278,7 @@ def main(args):
     if args.warmup:
         img = np.random.uniform(0, 255, [32, 320, 3]).astype(np.uint8)
         for i in range(2):
-            res = text_recognizer([img])
+            res = text_recognizer([img] * int(args.rec_batch_num))
 
     for image_file in image_file_list:
         img, flag = check_and_read_gif(image_file)

tools/infer/utility.py

@@ -159,6 +159,11 @@ def create_predictor(args, mode, logger):
         precision = inference.PrecisionType.Float32
 
     if args.use_gpu:
+        gpu_id = get_infer_gpuid()
+        if gpu_id is None:
+            raise ValueError(
+                "Not found GPU in current device. Please check your device or set args.use_gpu as False"
+            )
         config.enable_use_gpu(args.gpu_mem, 0)
         if args.use_tensorrt:
             config.enable_tensorrt_engine(