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

MANIFEST.in

-include LICENSE.txt
+include LICENSE
 include README.md
 
-recursive-include ppocr/utils *.txt utility.py logging.py
+recursive-include ppocr/utils *.txt utility.py logging.py network.py
 recursive-include ppocr/data/ *.py
 recursive-include ppocr/postprocess *.py
 recursive-include tools/infer *.py

configs/rec/ch_ppocr_v2.1/rec_chinese_lite_train_distillation_v2.1.yml

+Global:
+  debug: false
+  use_gpu: true
+  epoch_num: 800
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec_chinese_lite_distillation_v2.1
+  save_epoch_step: 3
+  eval_batch_step: [0, 2000]
+  cal_metric_during_train: true
+  pretrained_model:
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: false
+  infer_img: doc/imgs_words/ch/word_1.jpg
+  character_dict_path: ppocr/utils/ppocr_keys_v1.txt
+  character_type: ch
+  max_text_length: 25
+  infer_mode: false
+  use_space_char: false
+  distributed: true
+  save_res_path: ./output/rec/predicts_chinese_lite_distillation_v2.1.txt
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Cosine
+    learning_rate: 0.0005
+    warmup_epoch: 5
+  regularizer:
+    name: L2
+    factor: 1.0e-05
+Architecture:
+  name: DistillationModel
+  algorithm: Distillation
+  Models:
+    Student:
+      pretrained:
+      freeze_params: false
+      return_all_feats: true
+      model_type: rec
+      algorithm: CRNN
+      Transform:
+      Backbone:
+        name: MobileNetV3
+        scale: 0.5
+        model_name: small
+        small_stride: [1, 2, 2, 2]
+      Neck:
+        name: SequenceEncoder
+        encoder_type: rnn
+        hidden_size: 64
+      Head:
+        name: CTCHead
+        mid_channels: 96
+        fc_decay: 0.00001
+    Teacher:
+      pretrained:
+      freeze_params: false
+      return_all_feats: true
+      model_type: rec
+      algorithm: CRNN
+      Transform:
+      Backbone:
+        name: MobileNetV3
+        scale: 0.5
+        model_name: small
+        small_stride: [1, 2, 2, 2]
+      Neck:
+        name: SequenceEncoder
+        encoder_type: rnn
+        hidden_size: 64
+      Head:
+        name: CTCHead
+        mid_channels: 96
+        fc_decay: 0.00001
+  
+
+Loss:
+  name: CombinedLoss
+  loss_config_list:
+  - DistillationCTCLoss:
+      weight: 1.0
+      model_name_list: ["Student", "Teacher"]
+      key: head_out
+  - DistillationDMLLoss:
+      weight: 1.0
+      act: "softmax"
+      model_name_pairs:
+      - ["Student", "Teacher"]
+      key: head_out
+  - DistillationDistanceLoss:
+      weight: 1.0
+      mode: "l2"
+      model_name_pairs:
+      - ["Student", "Teacher"]
+      key: backbone_out
+
+PostProcess:
+  name: DistillationCTCLabelDecode
+  model_name: ["Student", "Teacher"]
+  key: head_out
+
+Metric:
+  name: DistillationMetric
+  base_metric_name: RecMetric
+  main_indicator: acc
+  key: "Student"
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/
+    label_file_list:
+    - ./train_data/train_list.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - RecAug:
+    - CTCLabelEncode:
+    - RecResizeImg:
+        image_shape: [3, 32, 320]
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+  loader:
+    shuffle: true
+    batch_size_per_card: 128
+    drop_last: true
+    num_sections: 1
+    num_workers: 8
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data
+    label_file_list:
+    - ./train_data/val_list.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - CTCLabelEncode:
+    - RecResizeImg:
+        image_shape: [3, 32, 320]
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+  loader:
+    shuffle: false
+    drop_last: false
+    batch_size_per_card: 128
+    num_workers: 8

configs/table/table_mv3.yml

+Global:
+  use_gpu: true
+  epoch_num: 50
+  log_smooth_window: 20
+  print_batch_step: 5
+  save_model_dir: ./output/table_mv3/
+  save_epoch_step: 5
+  # evaluation is run every 400 iterations after the 0th iteration
+  eval_batch_step: [0, 400]
+  cal_metric_during_train: True
+  pretrained_model: 
+  checkpoints: 
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_words/ch/word_1.jpg
+  # for data or label process
+  character_dict_path: ppocr/utils/dict/table_structure_dict.txt
+  character_type: en
+  max_text_length: 100
+  max_elem_length: 500
+  max_cell_num: 500
+  infer_mode: False
+  process_total_num: 0
+  process_cut_num: 0
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  clip_norm: 5.0
+  lr:
+    learning_rate: 0.001
+  regularizer:
+    name: 'L2'
+    factor: 0.00000
+
+Architecture:
+  model_type: table
+  algorithm: TableAttn
+  Backbone:
+    name: MobileNetV3
+    scale: 1.0
+    model_name: small
+    disable_se: True
+  Head:
+    name: TableAttentionHead
+    hidden_size: 256
+    l2_decay: 0.00001
+    loc_type: 2
+
+Loss:
+  name: TableAttentionLoss
+  structure_weight: 100.0
+  loc_weight: 10000.0
+
+PostProcess:
+  name: TableLabelDecode
+
+Metric:
+  name: TableMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: PubTabDataSet
+    data_dir: train_data/table/pubtabnet/train/
+    label_file_path: train_data/table/pubtabnet/PubTabNet_2.0.0_train.jsonl
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - ResizeTableImage:
+          max_len: 488
+      - TableLabelEncode:
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - PaddingTableImage:
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'structure', 'bbox_list', 'sp_tokens', 'bbox_list_mask']
+  loader:
+    shuffle: True
+    batch_size_per_card: 32
+    drop_last: True
+    num_workers: 1
+
+Eval:
+  dataset:
+    name: PubTabDataSet
+    data_dir: train_data/table/pubtabnet/val/
+    label_file_path: train_data/table/pubtabnet/PubTabNet_2.0.0_val.jsonl
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - ResizeTableImage:
+          max_len: 488
+      - TableLabelEncode:
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - PaddingTableImage:
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'structure', 'bbox_list', 'sp_tokens', 'bbox_list_mask']
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 16
+    num_workers: 1

deploy/android_demo/app/src/main/java/com/baidu/paddle/lite/demo/ocr/MainActivity.java

@@ -465,9 +465,13 @@ public class MainActivity extends AppCompatActivity {
     }
 
     public void btn_load_model_click(View view) {
+        if (predictor.isLoaded()){
+            tvStatus.setText("STATUS: model has been loaded");
+        }else{
             tvStatus.setText("STATUS: load model ......");
             loadModel();
         }
+    }
 
     public void btn_run_model_click(View view) {
         Bitmap image =((BitmapDrawable)ivInputImage.getDrawable()).getBitmap();

deploy/android_demo/app/src/main/java/com/baidu/paddle/lite/demo/ocr/Predictor.java

@@ -194,26 +194,25 @@ public class Predictor {
                         "supported!");
                 return false;
             }
+
             int[] channelStride = new int[]{width * height, width * height * 2};
-            int p = scaleImage.getPixel(scaleImage.getWidth() - 1, scaleImage.getHeight() - 1);
-            for (int y = 0; y < height; y++) {
-                for (int x = 0; x < width; x++) {
-                    int color = scaleImage.getPixel(x, y);
+            int[] pixels=new int[width*height];
+            scaleImage.getPixels(pixels,0,scaleImage.getWidth(),0,0,scaleImage.getWidth(),scaleImage.getHeight());
+            for (int i = 0; i < pixels.length; i++) {
+                int color = pixels[i];
                 float[] rgb = new float[]{(float) red(color) / 255.0f, (float) green(color) / 255.0f,
                         (float) blue(color) / 255.0f};
-                    inputData[y * width + x] = (rgb[channelIdx[0]] - inputMean[0]) / inputStd[0];
-                    inputData[y * width + x + channelStride[0]] = (rgb[channelIdx[1]] - inputMean[1]) / inputStd[1];
-                    inputData[y * width + x + channelStride[1]] = (rgb[channelIdx[2]] - inputMean[2]) / inputStd[2];
-
-                }
+                inputData[i] = (rgb[channelIdx[0]] - inputMean[0]) / inputStd[0];
+                inputData[i + channelStride[0]] = (rgb[channelIdx[1]] - inputMean[1]) / inputStd[1];
+                inputData[i+ channelStride[1]] = (rgb[channelIdx[2]] - inputMean[2]) / inputStd[2];
             }
         } else if (channels == 1) {
-            for (int y = 0; y < height; y++) {
-                for (int x = 0; x < width; x++) {
-                    int color = inputImage.getPixel(x, y);
+            int[] pixels=new int[width*height];
+            scaleImage.getPixels(pixels,0,scaleImage.getWidth(),0,0,scaleImage.getWidth(),scaleImage.getHeight());
+            for (int i = 0; i < pixels.length; i++) {
+                int color = pixels[i];
                 float gray = (float) (red(color) + green(color) + blue(color)) / 3.0f / 255.0f;
-                    inputData[y * width + x] = (gray - inputMean[0]) / inputStd[0];
-                }
+                inputData[i] = (gray - inputMean[0]) / inputStd[0];
             }
         } else {
             Log.i(TAG, "Unsupported channel size " + Integer.toString(channels) + ",  only channel 1 and 3 is " +

deploy/cpp_infer/include/utility.h

@@ -44,6 +44,9 @@ public:
   inline static size_t argmax(ForwardIterator first, ForwardIterator last) {
     return std::distance(first, std::max_element(first, last));
   }
+
+  static void GetAllFiles(const char *dir_name,
+                          std::vector<std::string> &all_inputs);
 };
 
 } // namespace PaddleOCR
\ No newline at end of file

deploy/cpp_infer/readme.md

@@ -77,7 +77,7 @@ opencv3/
 
 #### 1.2.1 直接下载安装
 
-* [Paddle预测库官网](https://www.paddlepaddle.org.cn/documentation/docs/zh/advanced_guide/inference_deployment/inference/build_and_install_lib_cn.html)上提供了不同cuda版本的Linux预测库，可以在官网查看并选择合适的预测库版本（*建议选择paddle版本>=2.0.1版本的预测库* ）。
+* [Paddle预测库官网](https://www.paddlepaddle.org.cn/documentation/docs/zh/2.0/guides/05_inference_deployment/inference/build_and_install_lib_cn.html) 上提供了不同cuda版本的Linux预测库，可以在官网查看并选择合适的预测库版本（*建议选择paddle版本>=2.0.1版本的预测库* ）。
 
 * 下载之后使用下面的方法解压。
 
@@ -89,10 +89,11 @@ tar -xf paddle_inference.tgz
 
 #### 1.2.2 预测库源码编译
 * 如果希望获取最新预测库特性，可以从Paddle github上克隆最新代码，源码编译预测库。
-* 可以参考[Paddle预测库官网](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/guides/05_inference_deployment/inference/build_and_install_lib_cn.html)的说明，从github上获取Paddle代码，然后进行编译，生成最新的预测库。使用git获取代码方法如下。
+* 可以参考[Paddle预测库安装编译说明](https://www.paddlepaddle.org.cn/documentation/docs/zh/2.0/guides/05_inference_deployment/inference/build_and_install_lib_cn.html#congyuanmabianyi) 的说明，从github上获取Paddle代码，然后进行编译，生成最新的预测库。使用git获取代码方法如下。
 
 ```shell
 git clone https://github.com/PaddlePaddle/Paddle.git
+git checkout release/2.1
 ```
 
 * 进入Paddle目录后，编译方法如下。
@@ -115,7 +116,7 @@ make -j
 make inference_lib_dist
 ```
 
-更多编译参数选项可以参考Paddle C++预测库官网：[https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/guides/05_inference_deployment/inference/build_and_install_lib_cn.html](https://www.paddlepaddle.org.cn/documentation/docs/zh/develop/guides/05_inference_deployment/inference/build_and_install_lib_cn.html)。
+更多编译参数选项介绍可以参考[文档说明](https://www.paddlepaddle.org.cn/documentation/docs/zh/2.0/guides/05_inference_deployment/inference/build_and_install_lib_cn.html#congyuanmabianyi)。
 
 
 * 编译完成之后，可以在`build/paddle_inference_install_dir/`文件下看到生成了以下文件及文件夹。
@@ -140,11 +141,11 @@ build/paddle_inference_install_dir/
 ```
 inference/
 |-- det_db
-|   |--inference.pdparams
-|   |--inference.pdimodel
+|   |--inference.pdiparams
+|   |--inference.pdmodel
 |-- rec_rcnn
-|   |--inference.pdparams
-|   |--inference.pdparams
+|   |--inference.pdiparams
+|   |--inference.pdmodel
 ```
 
 

deploy/cpp_infer/readme_en.md

@@ -78,8 +78,7 @@ opencv3/
 
 #### 1.2.1 Direct download and installation
 
-* Different cuda versions of the Linux inference library (based on GCC 4.8.2) are provided on the
-[Paddle inference library official website](https://www.paddlepaddle.org.cn/documentation/docs/en/develop/guides/05_inference_deployment/inference/build_and_install_lib_en.html). You can view and select the appropriate version of the inference library on the official website.
+[Paddle inference library official website](https://www.paddlepaddle.org.cn/documentation/docs/zh/2.0/guides/05_inference_deployment/inference/build_and_install_lib_cn.html). You can view and select the appropriate version of the inference library on the official website.
 
 
 * After downloading, use the following method to uncompress.
@@ -97,9 +96,10 @@ Finally you can see the following files in the folder of `paddle_inference/`.
 
 ```shell
 git clone https://github.com/PaddlePaddle/Paddle.git
+git checkout release/2.1
 ```
 
-* After entering the Paddle directory, the compilation method is as follows.
+* After entering the Paddle directory, the commands to compile the paddle inference library are as follows.
 
 ```shell
 rm -rf build
@@ -119,7 +119,7 @@ make -j
 make inference_lib_dist
 ```
 
-For more compilation parameter options, please refer to the official website of the Paddle C++ inference library:[https://www.paddlepaddle.org.cn/documentation/docs/en/develop/guides/05_inference_deployment/inference/build_and_install_lib_en.html](https://www.paddlepaddle.org.cn/documentation/docs/en/develop/guides/05_inference_deployment/inference/build_and_install_lib_en.html).
+For more compilation parameter options, please refer to the [document](https://www.paddlepaddle.org.cn/documentation/docs/zh/2.0/guides/05_inference_deployment/inference/build_and_install_lib_cn.html#congyuanmabianyi).
 
 
 * After the compilation process, you can see the following files in the folder of `build/paddle_inference_install_dir/`.
@@ -144,11 +144,11 @@ Among them, `paddle` is the Paddle library required for C++ prediction later, an
 ```
 inference/
 |-- det_db
-|   |--inference.pdparams
-|   |--inference.pdimodel
+|   |--inference.pdiparams
+|   |--inference.pdmodel
 |-- rec_rcnn
-|   |--inference.pdparams
-|   |--inference.pdparams
+|   |--inference.pdiparams
+|   |--inference.pdmodel
 ```
 
 

deploy/cpp_infer/src/main.cpp

@@ -27,9 +27,12 @@
 #include <fstream>
 #include <numeric>
 
+#include <glog/logging.h>
 #include <include/config.h>
 #include <include/ocr_det.h>
 #include <include/ocr_rec.h>
+#include <include/utility.h>
+#include <sys/stat.h>
 
 using namespace std;
 using namespace cv;
@@ -47,13 +50,8 @@ int main(int argc, char **argv) {
   config.PrintConfigInfo();
 
   std::string img_path(argv[2]);
-
-  cv::Mat srcimg = cv::imread(img_path, cv::IMREAD_COLOR);
-
-  if (!srcimg.data) {
-    std::cerr << "[ERROR] image read failed! image path: " << img_path << "\n";
-    exit(1);
-  }
+  std::vector<std::string> all_img_names;
+  Utility::GetAllFiles((char *)img_path.c_str(), all_img_names);
 
   DBDetector det(config.det_model_dir, config.use_gpu, config.gpu_id,
                  config.gpu_mem, config.cpu_math_library_num_threads,
@@ -76,7 +74,18 @@ int main(int argc, char **argv) {
                      config.use_tensorrt, config.use_fp16);
 
   auto start = std::chrono::system_clock::now();
+
+  for (auto img_dir : all_img_names) {
+    LOG(INFO) << "The predict img: " << img_dir;
+
+    cv::Mat srcimg = cv::imread(img_dir, cv::IMREAD_COLOR);
+    if (!srcimg.data) {
+      std::cerr << "[ERROR] image read failed! image path: " << img_path
+                << "\n";
+      exit(1);
+    }
     std::vector<std::vector<std::vector<int>>> boxes;
+
     det.Run(srcimg, boxes);
 
     rec.Run(boxes, srcimg, cls);
@@ -88,6 +97,7 @@ int main(int argc, char **argv) {
                      std::chrono::microseconds::period::num /
                      std::chrono::microseconds::period::den
               << "s" << std::endl;
+  }
 
   return 0;
 }

deploy/cpp_infer/src/ocr_det.cpp

@@ -30,6 +30,42 @@ void DBDetector::LoadModel(const std::string &model_dir) {
           this->use_fp16_ ? paddle_infer::Config::Precision::kHalf
                           : paddle_infer::Config::Precision::kFloat32,
           false, false);
+      std::map<std::string, std::vector<int>> min_input_shape = {
+          {"x", {1, 3, 50, 50}},
+          {"conv2d_92.tmp_0", {1, 96, 20, 20}},
+          {"conv2d_91.tmp_0", {1, 96, 10, 10}},
+          {"nearest_interp_v2_1.tmp_0", {1, 96, 10, 10}},
+          {"nearest_interp_v2_2.tmp_0", {1, 96, 20, 20}},
+          {"nearest_interp_v2_3.tmp_0", {1, 24, 20, 20}},
+          {"nearest_interp_v2_4.tmp_0", {1, 24, 20, 20}},
+          {"nearest_interp_v2_5.tmp_0", {1, 24, 20, 20}},
+          {"elementwise_add_7", {1, 56, 2, 2}},
+          {"nearest_interp_v2_0.tmp_0", {1, 96, 2, 2}}};
+      std::map<std::string, std::vector<int>> max_input_shape = {
+          {"x", {1, 3, this->max_side_len_, this->max_side_len_}},
+          {"conv2d_92.tmp_0", {1, 96, 400, 400}},
+          {"conv2d_91.tmp_0", {1, 96, 200, 200}},
+          {"nearest_interp_v2_1.tmp_0", {1, 96, 200, 200}},
+          {"nearest_interp_v2_2.tmp_0", {1, 96, 400, 400}},
+          {"nearest_interp_v2_3.tmp_0", {1, 24, 400, 400}},
+          {"nearest_interp_v2_4.tmp_0", {1, 24, 400, 400}},
+          {"nearest_interp_v2_5.tmp_0", {1, 24, 400, 400}},
+          {"elementwise_add_7", {1, 56, 400, 400}},
+          {"nearest_interp_v2_0.tmp_0", {1, 96, 400, 400}}};
+      std::map<std::string, std::vector<int>> opt_input_shape = {
+          {"x", {1, 3, 640, 640}},
+          {"conv2d_92.tmp_0", {1, 96, 160, 160}},
+          {"conv2d_91.tmp_0", {1, 96, 80, 80}},
+          {"nearest_interp_v2_1.tmp_0", {1, 96, 80, 80}},
+          {"nearest_interp_v2_2.tmp_0", {1, 96, 160, 160}},
+          {"nearest_interp_v2_3.tmp_0", {1, 24, 160, 160}},
+          {"nearest_interp_v2_4.tmp_0", {1, 24, 160, 160}},
+          {"nearest_interp_v2_5.tmp_0", {1, 24, 160, 160}},
+          {"elementwise_add_7", {1, 56, 40, 40}},
+          {"nearest_interp_v2_0.tmp_0", {1, 96, 40, 40}}};
+
+      config.SetTRTDynamicShapeInfo(min_input_shape, max_input_shape,
+                                    opt_input_shape);
     }
   } else {
     config.DisableGpu();
@@ -48,7 +84,7 @@ void DBDetector::LoadModel(const std::string &model_dir) {
   config.SwitchIrOptim(true);
 
   config.EnableMemoryOptim();
-  config.DisableGlogInfo();
+  // config.DisableGlogInfo();
 
   this->predictor_ = CreatePredictor(config);
 }

deploy/cpp_infer/src/ocr_rec.cpp

@@ -106,6 +106,15 @@ void CRNNRecognizer::LoadModel(const std::string &model_dir) {
           this->use_fp16_ ? paddle_infer::Config::Precision::kHalf
                           : paddle_infer::Config::Precision::kFloat32,
           false, false);
+      std::map<std::string, std::vector<int>> min_input_shape = {
+          {"x", {1, 3, 32, 10}}};
+      std::map<std::string, std::vector<int>> max_input_shape = {
+          {"x", {1, 3, 32, 2000}}};
+      std::map<std::string, std::vector<int>> opt_input_shape = {
+          {"x", {1, 3, 32, 320}}};
+
+      config.SetTRTDynamicShapeInfo(min_input_shape, max_input_shape,
+                                    opt_input_shape);
     }
   } else {
     config.DisableGpu();

deploy/cpp_infer/src/preprocess_op.cpp

@@ -47,16 +47,13 @@ void Normalize::Run(cv::Mat *im, const std::vector<float> &mean,
     e /= 255.0;
   }
   (*im).convertTo(*im, CV_32FC3, e);
-  for (int h = 0; h < im->rows; h++) {
-    for (int w = 0; w < im->cols; w++) {
-      im->at<cv::Vec3f>(h, w)[0] =
-          (im->at<cv::Vec3f>(h, w)[0] - mean[0]) * scale[0];
-      im->at<cv::Vec3f>(h, w)[1] =
-          (im->at<cv::Vec3f>(h, w)[1] - mean[1]) * scale[1];
-      im->at<cv::Vec3f>(h, w)[2] =
-          (im->at<cv::Vec3f>(h, w)[2] - mean[2]) * scale[2];
-    }
+  std::vector<cv::Mat> bgr_channels(3);
+  cv::split(*im, bgr_channels);
+  for (auto i = 0; i < bgr_channels.size(); i++) {
+    bgr_channels[i].convertTo(bgr_channels[i], CV_32FC1, 1.0 * scale[i],
+                              (0.0 - mean[i]) * scale[i]);
   }
+  cv::merge(bgr_channels, *im);
 }
 
 void ResizeImgType0::Run(const cv::Mat &img, cv::Mat &resize_img,
@@ -81,15 +78,9 @@ void ResizeImgType0::Run(const cv::Mat &img, cv::Mat &resize_img,
   resize_h = max(int(round(float(resize_h) / 32) * 32), 32);
   resize_w = max(int(round(float(resize_w) / 32) * 32), 32);
 
-  if (!use_tensorrt) {
   cv::resize(img, resize_img, cv::Size(resize_w, resize_h));
   ratio_h = float(resize_h) / float(h);
   ratio_w = float(resize_w) / float(w);
-  } else {
-    cv::resize(img, resize_img, cv::Size(640, 640));
-    ratio_h = float(640) / float(h);
-    ratio_w = float(640) / float(w);
-  }
 }
 
 void CrnnResizeImg::Run(const cv::Mat &img, cv::Mat &resize_img, float wh_ratio,
@@ -108,23 +99,12 @@ void CrnnResizeImg::Run(const cv::Mat &img, cv::Mat &resize_img, float wh_ratio,
     resize_w = imgW;
   else
     resize_w = int(ceilf(imgH * ratio));
-  if (!use_tensorrt) {
+
   cv::resize(img, resize_img, cv::Size(resize_w, imgH), 0.f, 0.f,
              cv::INTER_LINEAR);
   cv::copyMakeBorder(resize_img, resize_img, 0, 0, 0,
                      int(imgW - resize_img.cols), cv::BORDER_CONSTANT,
                      {127, 127, 127});
-  } else {
-    int k = int(img.cols * 32 / img.rows);
-    if (k >= 100) {
-      cv::resize(img, resize_img, cv::Size(100, 32), 0.f, 0.f,
-                 cv::INTER_LINEAR);
-    } else {
-      cv::resize(img, resize_img, cv::Size(k, 32), 0.f, 0.f, cv::INTER_LINEAR);
-      cv::copyMakeBorder(resize_img, resize_img, 0, 0, 0, int(100 - k),
-                         cv::BORDER_CONSTANT, {127, 127, 127});
-    }
-  }
 }
 
 void ClsResizeImg::Run(const cv::Mat &img, cv::Mat &resize_img,
@@ -142,16 +122,12 @@ void ClsResizeImg::Run(const cv::Mat &img, cv::Mat &resize_img,
   else
     resize_w = int(ceilf(imgH * ratio));
 
-  if (!use_tensorrt) {
   cv::resize(img, resize_img, cv::Size(resize_w, imgH), 0.f, 0.f,
              cv::INTER_LINEAR);
   if (resize_w < imgW) {
     cv::copyMakeBorder(resize_img, resize_img, 0, 0, 0, imgW - resize_w,
                        cv::BORDER_CONSTANT, cv::Scalar(0, 0, 0));
   }
-  } else {
-    cv::resize(img, resize_img, cv::Size(100, 32), 0.f, 0.f, cv::INTER_LINEAR);
-  }
 }
 
 } // namespace PaddleOCR

deploy/cpp_infer/src/utility.cpp

@@ -12,12 +12,14 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+#include <dirent.h>
+#include <include/utility.h>
 #include <iostream>
 #include <ostream>
+#include <sys/stat.h>
+#include <sys/types.h>
 #include <vector>
 
-#include <include/utility.h>
-
 namespace PaddleOCR {
 
 std::vector<std::string> Utility::ReadDict(const std::string &path) {
@@ -57,4 +59,37 @@ void Utility::VisualizeBboxes(
             << std::endl;
 }
 
+// list all files under a directory
+void Utility::GetAllFiles(const char *dir_name,
+                          std::vector<std::string> &all_inputs) {
+  if (NULL == dir_name) {
+    std::cout << " dir_name is null ! " << std::endl;
+    return;
+  }
+  struct stat s;
+  lstat(dir_name, &s);
+  if (!S_ISDIR(s.st_mode)) {
+    std::cout << "dir_name is not a valid directory !" << std::endl;
+    all_inputs.push_back(dir_name);
+    return;
+  } else {
+    struct dirent *filename; // return value for readdir()
+    DIR *dir;                // return value for opendir()
+    dir = opendir(dir_name);
+    if (NULL == dir) {
+      std::cout << "Can not open dir " << dir_name << std::endl;
+      return;
+    }
+    std::cout << "Successfully opened the dir !" << std::endl;
+    while ((filename = readdir(dir)) != NULL) {
+      if (strcmp(filename->d_name, ".") == 0 ||
+          strcmp(filename->d_name, "..") == 0)
+        continue;
+      // img_dir + std::string("/") + all_inputs[0];
+      all_inputs.push_back(dir_name + std::string("/") +
+                           std::string(filename->d_name));
+    }
+  }
+}
+
 } // namespace PaddleOCR
\ No newline at end of file

deploy/cpp_infer/tools/build.sh

@@ -12,9 +12,10 @@ cmake .. \
     -DWITH_MKL=ON \
     -DWITH_GPU=OFF \
     -DWITH_STATIC_LIB=OFF \
-    -DUSE_TENSORRT=OFF \
+    -DWITH_TENSORRT=OFF \
     -DOPENCV_DIR=${OPENCV_DIR} \
     -DCUDNN_LIB=${CUDNN_LIB_DIR} \
     -DCUDA_LIB=${CUDA_LIB_DIR} \
+    -DTENSORRT_DIR=${TENSORRT_DIR} \
 
 make -j

deploy/cpp_infer/tools/config.txt

@@ -23,7 +23,7 @@ rec_model_dir  ./inference/ch_ppocr_mobile_v2.0_rec_infer/
 char_list_file  ../../ppocr/utils/ppocr_keys_v1.txt
 
 # show the detection results
-visualize 1
+visualize 0
 
 # use_tensorrt
 use_tensorrt 0

deploy/hubserving/readme.md

@@ -29,7 +29,7 @@ deploy/hubserving/ocr_system/
 ### 1. 准备环境
 ```shell
 # 安装paddlehub  
-pip3 install paddlehub --upgrade -i https://pypi.tuna.tsinghua.edu.cn/simple
+pip3 install paddlehub==1.8.3 --upgrade -i https://pypi.tuna.tsinghua.edu.cn/simple
 ```
 
 ### 2. 下载推理模型

deploy/hubserving/readme_en.md

@@ -30,7 +30,7 @@ The following steps take the 2-stage series service as an example. If only the d
 ### 1. Prepare the environment
 ```shell
 # Install paddlehub  
-pip3 install paddlehub --upgrade -i https://pypi.tuna.tsinghua.edu.cn/simple
+pip3 install paddlehub==1.8.3 --upgrade -i https://pypi.tuna.tsinghua.edu.cn/simple
 ```
 
 ### 2. Download inference model

doc/doc_ch/config.md

@@ -111,9 +111,9 @@
 |         字段             |            用途            |      默认值        |            备注             |
 | :---------------------: |  :---------------------:   | :--------------:  |   :--------------------:   |
 |      **dataset**        |         每次迭代返回一个样本          |  -  |  -  |
-|      name        |        dataset类名         |  SimpleDataSet |  目前支持`SimpleDataSet`和`LMDBDateSet`  |
+|      name        |        dataset类名         |  SimpleDataSet |  目前支持`SimpleDataSet`和`LMDBDataSet`  |
 |      data_dir        |        数据集图片存放路径         |  ./train_data |  \  |
-|      label_file_list        |        数据标签路径         |  ["./train_data/train_list.txt"] | dataset为LMDBDateSet时不需要此参数   |
+|      label_file_list        |        数据标签路径         |  ["./train_data/train_list.txt"] | dataset为LMDBDataSet时不需要此参数   |
 |      ratio_list        |        数据集的比例         |  [1.0] | 若label_file_list中有两个train_list，且ratio_list为[0.4,0.6]，则从train_list1中采样40%，从train_list2中采样60%组合整个dataset   |
 |      transforms        |        对图片和标签进行变换的方法列表         |  [DecodeImage,CTCLabelEncode,RecResizeImg,KeepKeys] |   见[ppocr/data/imaug](../../ppocr/data/imaug)  |
 |      **loader**        |        dataloader相关         |  - |   |

doc/doc_ch/recognition.md

@@ -243,7 +243,7 @@ Optimizer:
 
 Train:
   dataset:
-    # 数据集格式，支持LMDBDateSet以及SimpleDataSet
+    # 数据集格式，支持LMDBDataSet以及SimpleDataSet
     name: SimpleDataSet
     # 数据集路径
     data_dir: ./train_data/
@@ -263,7 +263,7 @@ Train:
 
 Eval:
   dataset:
-    # 数据集格式，支持LMDBDateSet以及SimpleDataSet
+    # 数据集格式，支持LMDBDataSet以及SimpleDataSet
     name: SimpleDataSet
     # 数据集路径
     data_dir: ./train_data
@@ -393,7 +393,7 @@ Global:
 
 Train:
   dataset:
-    # 数据集格式，支持LMDBDateSet以及SimpleDataSet
+    # 数据集格式，支持LMDBDataSet以及SimpleDataSet
     name: SimpleDataSet
     # 数据集路径
     data_dir: ./train_data/
@@ -403,7 +403,7 @@ Train:
 
 Eval:
   dataset:
-    # 数据集格式，支持LMDBDateSet以及SimpleDataSet
+    # 数据集格式，支持LMDBDataSet以及SimpleDataSet
     name: SimpleDataSet
     # 数据集路径
     data_dir: ./train_data

doc/doc_ch/whl.md

@@ -59,7 +59,7 @@ im_show.save('result.jpg')
 from paddleocr import PaddleOCR, draw_ocr
 ocr = PaddleOCR() # need to run only once to download and load model into memory
 img_path = 'PaddleOCR/doc/imgs/11.jpg'
-result = ocr.ocr(img_path)
+result = ocr.ocr(img_path,cls=False)
 for line in result:
     print(line)
 
@@ -355,3 +355,4 @@ im_show.save('result.jpg')
 | det                     | 前向时使用启动检测                                                                                                                                                                                                   | TRUE                    |
 | rec                     | 前向时是否启动识别                                                                                                                                                                                                   | TRUE                    |
 | cls                     | 前向时是否启动分类 (命令行模式下使用use_angle_cls控制前向是否启动分类)                                                                                                                                                                                                | FALSE                    |
+| show_log                     | 是否打印det和rec等信息                                                                                                                                                                                                | FALSE                    |