Merge pull request #989 from WenmuZhou/dygraph

yml文件删掉一些个人路径

configs/det/det_mv3_db.yml

@@ -10,8 +10,8 @@ Global:
   # if pretrained_model is saved in static mode, load_static_weights must set to True
   load_static_weights: True
   cal_metric_during_train: False
-  pretrained_model: /home/zhoujun20/pretrain_models/MobileNetV3_large_x0_5_pretrained
+  pretrained_model: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
-  checkpoints: #./output/det_db_0.001_DiceLoss_256_pp_config_2.0b_4gpu/best_accuracy
+  checkpoints:
   save_inference_dir:
   use_visualdl: True
   infer_img: doc/imgs_en/img_10.jpg
@@ -22,9 +22,7 @@ Optimizer:
   beta1: 0.9
   beta2: 0.999
   learning_rate:
-#    name: Cosine
     lr: 0.001
-#    warmup_epoch: 0
   regularizer:
     name: 'L2'
     factor: 0
@@ -98,7 +96,7 @@ TRAIN:
           order: 'hwc'
       - ToCHWImage:
       - keepKeys:
-          keep_keys: ['image','threshold_map','threshold_mask','shrink_map','shrink_mask'] # dataloader将按照此顺序返回list
+          keep_keys: ['image','threshold_map','threshold_mask','shrink_map','shrink_mask'] # dataloader will return list in this order
   loader:
     shuffle: True
     drop_last: False

configs/det/det_r50_vd_db.yml

@@ -3,15 +3,15 @@ Global:
   epoch_num: 1200
   log_smooth_window: 20
   print_batch_step: 2
-  save_model_dir: ./output/20201015_r50/
+  save_model_dir: ./output/det_r50_vd/
   save_epoch_step: 1200
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: 8
   # if pretrained_model is saved in static mode, load_static_weights must set to True
   load_static_weights: True
   cal_metric_during_train: False
-  pretrained_model: /home/zhoujun20/pretrain_models/ResNet50_vd_ssld_pretrained/
+  pretrained_model: ./pretrain_models/ResNet50_vd_ssld_pretrained/
-  checkpoints: #./output/det_db_0.001_DiceLoss_256_pp_config_2.0b_4gpu/best_accuracy
+  checkpoints:
   save_inference_dir:
   use_visualdl: True
   infer_img: doc/imgs_en/img_10.jpg
@@ -22,9 +22,7 @@ Optimizer:
   beta1: 0.9
   beta2: 0.999
   learning_rate:
-#    name: Cosine
     lr: 0.001
-#    warmup_epoch: 0
   regularizer:
     name: 'L2'
     factor: 0
@@ -65,9 +63,9 @@ Metric:
 TRAIN:
   dataset:
     name: SimpleDataSet
-    data_dir: /home/zhoujun20/detection/
+    data_dir: ./detection/
     file_list:
-      - /home/zhoujun20/detection/train_icdar2015_label.txt # dataset1
+      - ./detection/train_icdar2015_label.txt # dataset1
     ratio_list: [1.0]
     transforms:
       - DecodeImage: # load image
@@ -97,7 +95,7 @@ TRAIN:
           order: 'hwc'
       - ToCHWImage:
       - keepKeys:
-          keep_keys: ['image','threshold_map','threshold_mask','shrink_map','shrink_mask'] # dataloader将按照此顺序返回list
+          keep_keys: ['image','threshold_map','threshold_mask','shrink_map','shrink_mask'] # dataloader will return list in this order
   loader:
     shuffle: True
     drop_last: False
@@ -107,9 +105,9 @@ TRAIN:
 EVAL:
   dataset:
     name: SimpleDataSet
-    data_dir: /home/zhoujun20/detection/
+    data_dir: ./detection/
     file_list:
-      - /home/zhoujun20/detection/test_icdar2015_label.txt
+      - ./detection/test_icdar2015_label.txt
     transforms:
       - DecodeImage: # load image
           img_mode: BGR

configs/rec/rec_mv3_none_bilstm_ctc.yml

@@ -3,7 +3,7 @@ Global:
   epoch_num: 500
   log_smooth_window: 20
   print_batch_step: 10
-  save_model_dir: ./output/rec/test/
+  save_model_dir: ./output/rec/mv3_none_bilstm_ctc/
   save_epoch_step: 500
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: 127
@@ -11,7 +11,7 @@ Global:
   load_static_weights: True
   cal_metric_during_train: True
   pretrained_model:
-  checkpoints: #output/rec/rec_crnn/best_accuracy
+  checkpoints:
   save_inference_dir:
   use_visualdl: False
   infer_img: doc/imgs_words/ch/word_1.jpg
@@ -29,9 +29,7 @@ Optimizer:
   beta1: 0.9
   beta2: 0.999
   learning_rate:
-    name: Cosine
     lr: 0.001
-    warmup_epoch: 4
   regularizer:
     name: 'L2'
     factor: 0.00001
@@ -66,9 +64,9 @@ Metric:
 TRAIN:
   dataset:
     name: SimpleDataSet
-    data_dir: /home/zhoujun20/rec
+    data_dir: ./rec
     file_list:
-      - /home/zhoujun20/rec/real_data.txt # dataset1
+      - ./rec/train.txt # dataset1
     ratio_list: [ 0.4,0.6 ]
     transforms:
       - DecodeImage: # load image
@@ -79,7 +77,7 @@ TRAIN:
       - RecResizeImg:
           image_shape: [ 3,32,320 ]
       - keepKeys:
-          keep_keys: [ 'image','label','length' ] # dataloader将按照此顺序返回list
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
   loader:
     batch_size: 256
     shuffle: True
@@ -89,9 +87,9 @@ TRAIN:
 EVAL:
   dataset:
     name: SimpleDataSet
-    data_dir: /home/zhoujun20/rec
+    data_dir: ./rec
     file_list:
-      - /home/zhoujun20/rec/label_val_all.txt
+      - ./rec/val.txt
     transforms:
       - DecodeImage: # load image
           img_mode: BGR
@@ -100,7 +98,7 @@ EVAL:
       - RecResizeImg:
           image_shape: [ 3,32,320 ]
       - keepKeys:
-          keep_keys: [ 'image','label','length' ] # dataloader将按照此顺序返回list
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
   loader:
     shuffle: False
     drop_last: False

configs/rec/rec_mv3_none_bilstm_ctc_lmdb.yml

@@ -3,7 +3,7 @@ Global:
   epoch_num: 500
   log_smooth_window: 20
   print_batch_step: 1
-  save_model_dir: ./output/rec/test/
+  save_model_dir: ./output/rec/mv3_none_bilstm_ctc/
   save_epoch_step: 500
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: 1016
@@ -11,13 +11,13 @@ Global:
   load_static_weights: True
   cal_metric_during_train: True
   pretrained_model:
-  checkpoints: #output/rec/rec_crnn/best_accuracy
+  checkpoints:
   save_inference_dir:
   use_visualdl: True
   infer_img: doc/imgs_words/ch/word_1.jpg
   # for data or label process
   max_text_length: 80
-  character_dict_path: /home/zhoujun20/rec/lmdb/dict.txt
+  character_dict_path: ppocr/utils/ppocr_keys_v1.txt
   character_type: 'ch'
   use_space_char: True
   infer_mode: False
@@ -29,9 +29,7 @@ Optimizer:
   beta1: 0.9
   beta2: 0.999
   learning_rate:
-    name: Cosine
     lr: 0.0005
-    warmup_epoch: 1
   regularizer:
     name: 'L2'
     factor: 0.00001
@@ -67,7 +65,7 @@ TRAIN:
   dataset:
     name: LMDBDateSet
     file_list:
-      - /home/zhoujun20/rec/lmdb/train # dataset1
+      - ./rec/lmdb/train # dataset1
     ratio_list: [ 0.4,0.6 ]
     transforms:
       - DecodeImage: # load image
@@ -78,7 +76,7 @@ TRAIN:
       - RecResizeImg:
           image_shape: [ 3,32,320 ]
       - keepKeys:
-          keep_keys: [ 'image','label','length' ] # dataloader将按照此顺序返回list
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
   loader:
     batch_size: 256
     shuffle: True
@@ -89,7 +87,7 @@ EVAL:
   dataset:
     name: LMDBDateSet
     file_list:
-      - /home/zhoujun20/rec/lmdb/val
+      - ./rec/lmdb/val
     transforms:
       - DecodeImage: # load image
           img_mode: BGR
@@ -98,7 +96,7 @@ EVAL:
       - RecResizeImg:
           image_shape: [ 3,32,320 ]
       - keepKeys:
-          keep_keys: [ 'image','label','length' ] # dataloader将按照此顺序返回list
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
   loader:
     shuffle: False
     drop_last: False

configs/rec/rec_mv3_none_none_ctc_lmdb.yml

@@ -1,25 +1,25 @@
 Global:
   use_gpu: false
-  epoch_num: 500
+  epoch_num: 72
   log_smooth_window: 20
-  print_batch_step: 1
+  print_batch_step: 10
-  save_model_dir: ./output/rec/test/
+  save_model_dir: ./output/rec/mv3_none_none_ctc/
   save_epoch_step: 500
   # evaluation is run every 5000 iterations after the 4000th iteration
-  eval_batch_step: 1016
+  eval_batch_step: 2000
   # if pretrained_model is saved in static mode, load_static_weights must set to True
   load_static_weights: True
   cal_metric_during_train: True
   pretrained_model:
-  checkpoints: #output/rec/rec_crnn/best_accuracy
+  checkpoints:
   save_inference_dir:
   use_visualdl: True
   infer_img: doc/imgs_words/ch/word_1.jpg
   # for data or label process
-  max_text_length: 80
+  max_text_length: 25
-  character_dict_path: /home/zhoujun20/rec/lmdb/dict.txt
+  character_dict_path:
   character_type: 'en'
-  use_space_char: True
+  use_space_char: False
   infer_mode: False
   use_tps: False
 
@@ -29,9 +29,7 @@ Optimizer:
   beta1: 0.9
   beta2: 0.999
   learning_rate:
-    name: Cosine
     lr: 0.0005
-    warmup_epoch: 1
   regularizer:
     name: 'L2'
     factor: 0.00001
@@ -43,7 +41,7 @@ Architecture:
   Backbone:
     name: MobileNetV3
     scale: 0.5
-    model_name: small
+    model_name: large
     small_stride: [ 1, 2, 2, 2 ]
   Neck:
     name: SequenceEncoder
@@ -66,7 +64,7 @@ TRAIN:
   dataset:
     name: LMDBDateSet
     file_list:
-      - /Users/zhoujun20/Downloads/evaluation_new # dataset1
+      - ./rec/train # dataset1
     ratio_list: [ 0.4,0.6 ]
     transforms:
       - DecodeImage: # load image
@@ -75,9 +73,9 @@ TRAIN:
       - CTCLabelEncode: # Class handling label
       - RecAug:
       - RecResizeImg:
-          image_shape: [ 3,32,320 ]
+          image_shape: [ 3,32,100 ]
       - keepKeys:
-          keep_keys: [ 'image','label','length' ] # dataloader将按照此顺序返回list
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
   loader:
     batch_size: 256
     shuffle: True
@@ -88,16 +86,16 @@ EVAL:
   dataset:
     name: LMDBDateSet
     file_list:
-      - /home/zhoujun20/rec/lmdb/val
+      - ./rec/val/
     transforms:
       - DecodeImage: # load image
           img_mode: BGR
           channel_first: False
       - CTCLabelEncode: # Class handling label
       - RecResizeImg:
-          image_shape: [ 3,32,320 ]
+          image_shape: [ 3,32,100 ]
       - keepKeys:
-          keep_keys: [ 'image','label','length' ] # dataloader将按照此顺序返回list
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
   loader:
     shuffle: False
     drop_last: False

configs/rec/rec_r34_vd_none_bilstm_ctc.yml

@@ -3,7 +3,7 @@ Global:
   epoch_num: 500
   log_smooth_window: 20
   print_batch_step: 10
-  save_model_dir: ./output/rec/test/
+  save_model_dir: ./output/rec/res34_none_bilstm_ctc/
   save_epoch_step: 500
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: 127
@@ -11,7 +11,7 @@ Global:
   load_static_weights: True
   cal_metric_during_train: True
   pretrained_model:
-  checkpoints: #output/rec/rec_crnn/best_accuracy
+  checkpoints:
   save_inference_dir:
   use_visualdl: False
   infer_img: doc/imgs_words/ch/word_1.jpg
@@ -29,9 +29,7 @@ Optimizer:
   beta1: 0.9
   beta2: 0.999
   learning_rate:
-    name: Cosine
     lr: 0.001
-    warmup_epoch: 4
   regularizer:
     name: 'L2'
     factor: 0.00001
@@ -64,9 +62,9 @@ Metric:
 TRAIN:
   dataset:
     name: SimpleDataSet
-    data_dir: /home/zhoujun20/rec
+    data_dir: ./rec
     file_list:
-      - /home/zhoujun20/rec/real_data.txt # dataset1
+      - ./rec/train.txt # dataset1
     ratio_list: [ 0.4,0.6 ]
     transforms:
       - DecodeImage: # load image
@@ -77,7 +75,7 @@ TRAIN:
       - RecResizeImg:
           image_shape: [ 3,32,320 ]
       - keepKeys:
-          keep_keys: [ 'image','label','length' ] # dataloader将按照此顺序返回list
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
   loader:
     batch_size: 256
     shuffle: True
@@ -87,9 +85,9 @@ TRAIN:
 EVAL:
   dataset:
     name: SimpleDataSet
-    data_dir: /home/zhoujun20/rec
+    data_dir: ./rec
     file_list:
-      - /home/zhoujun20/rec/label_val_all.txt
+      - ./rec/val.txt
     transforms:
       - DecodeImage: # load image
           img_mode: BGR
@@ -98,7 +96,7 @@ EVAL:
       - RecResizeImg:
           image_shape: [ 3,32,320 ]
       - keepKeys:
-          keep_keys: [ 'image','label','length' ] # dataloader将按照此顺序返回list
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
   loader:
     shuffle: False
     drop_last: False

configs/rec/rec_r34_vd_none_none_ctc.yml

@@ -0,0 +1,103 @@
+Global:
+  use_gpu: false
+  epoch_num: 500
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec/res34_none_none_ctc/
+  save_epoch_step: 500
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: 127
+  # if pretrained_model is saved in static mode, load_static_weights must set to True
+  load_static_weights: True
+  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
+  max_text_length: 80
+  character_dict_path: ppocr/utils/ppocr_keys_v1.txt
+  character_type: 'ch'
+  use_space_char: False
+  infer_mode: False
+  use_tps: False
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  learning_rate:
+    lr: 0.001
+  regularizer:
+    name: 'L2'
+    factor: 0.00001
+
+Architecture:
+  type: rec
+  algorithm: CRNN
+  Transform:
+  Backbone:
+    name: ResNet
+    layers: 34
+  Neck:
+    name: SequenceEncoder
+    encoder_type: reshape
+  Head:
+    name: CTC
+    fc_decay: 0.00001
+
+Loss:
+  name: CTCLoss
+
+PostProcess:
+  name: CTCLabelDecode
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+
+TRAIN:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./rec
+    file_list:
+      - ./rec/train.txt # dataset1
+    ratio_list: [ 0.4,0.6 ]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - CTCLabelEncode: # Class handling label
+      - RecAug:
+      - RecResizeImg:
+          image_shape: [ 3,32,320 ]
+      - keepKeys:
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
+  loader:
+    batch_size: 256
+    shuffle: True
+    drop_last: True
+    num_workers: 8
+
+EVAL:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./rec
+    file_list:
+      - ./rec/val.txt
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - CTCLabelEncode: # Class handling label
+      - RecResizeImg:
+          image_shape: [ 3,32,320 ]
+      - keepKeys:
+          keep_keys: [ 'image','label','length' ] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size: 256
+    num_workers: 8

ppocr/modeling/architectures/model.py

@@ -21,7 +21,6 @@ __dir__ = os.path.dirname(os.path.abspath(__file__))
 sys.path.append(__dir__)
 sys.path.append('/home/zhoujun20/PaddleOCR')
 
-import paddle
 from paddle import nn
 from ppocr.modeling.transform import build_transform
 from ppocr.modeling.backbones import build_backbone
@@ -72,12 +71,10 @@ class Model(nn.Layer):
             config['Neck']['in_channels'] = in_channels
             self.neck = build_neck(config['Neck'])
             in_channels = self.neck.out_channels
-
+        # # build head, head is need for det, rec and cls
-        # # build head, head is need for del, rec and cls
         config["Head"]['in_channels'] = in_channels
         self.head = build_head(config["Head"])
 
-    # @paddle.jit.to_static
     def forward(self, x):
         if self.use_transform:
             x = self.transform(x)
@@ -85,41 +82,4 @@ class Model(nn.Layer):
         if self.use_neck:
             x = self.neck(x)
         x = self.head(x)
-        return x
+        return x
-
-
-def check_static():
-    import numpy as np
-    from ppocr.utils.save_load import load_dygraph_pretrain
-    from ppocr.utils.logging import get_logger
-    from tools import program
-
-    config = program.load_config('configs/rec/rec_r34_vd_none_bilstm_ctc.yml')
-
-    logger = get_logger()
-    np.random.seed(0)
-    data = np.random.rand(1, 3, 32, 320).astype(np.float32)
-    paddle.disable_static()
-
-    config['Architecture']['in_channels'] = 3
-    config['Architecture']["Head"]['out_channels'] = 6624
-    model = Model(config['Architecture'])
-    model.eval()
-    load_dygraph_pretrain(
-        model,
-        logger,
-        '/Users/zhoujun20/Desktop/code/PaddleOCR/cnn_ctc/cnn_ctc',
-        load_static_weights=True)
-    x = paddle.to_tensor(data)
-    y = model(x)
-    for y1 in y:
-        print(y1.shape)
-
-    static_out = np.load(
-        '/Users/zhoujun20/Desktop/code/PaddleOCR/output/conv.npy')
-    diff = y.numpy() - static_out
-    print(y.shape, static_out.shape, diff.mean())
-
-
-if __name__ == '__main__':
-    check_static()

ppocr/modeling/heads/rec_ctc_head.py

@@ -20,6 +20,7 @@ import math
 
 import paddle
 from paddle import ParamAttr, nn
+from paddle.nn import functional as F
 
 
 def get_para_bias_attr(l2_decay, k, name):
@@ -48,4 +49,6 @@ class CTC(nn.Layer):
 
     def forward(self, x, labels=None):
         predicts = self.fc(x)
+        if not self.training:
+            predicts = F.softmax(predicts, axis=2)
         return predicts

tools/infer/utility.py

@@ -14,11 +14,14 @@
 
 import argparse
 import os
+import sys
 import cv2
 import numpy as np
 import json
 from PIL import Image, ImageDraw, ImageFont
 import math
+from paddle.fluid.core import AnalysisConfig
+from paddle.fluid.core import create_paddle_predictor
 
 
 def parse_args():
@@ -71,6 +74,59 @@ def parse_args():
     return parser.parse_args()
 
 
+def create_predictor(args, mode, logger):
+    if mode == "det":
+        model_dir = args.det_model_dir
+    elif mode == 'cls':
+        model_dir = args.cls_model_dir
+    else:
+        model_dir = args.rec_model_dir
+
+    if model_dir is None:
+        logger.info("not find {} model file path {}".format(mode, model_dir))
+        sys.exit(0)
+    model_file_path = model_dir + "/model"
+    params_file_path = model_dir + "/params"
+    if not os.path.exists(model_file_path):
+        logger.info("not find model file path {}".format(model_file_path))
+        sys.exit(0)
+    if not os.path.exists(params_file_path):
+        logger.info("not find params file path {}".format(params_file_path))
+        sys.exit(0)
+
+    config = AnalysisConfig(model_file_path, params_file_path)
+
+    if args.use_gpu:
+        config.enable_use_gpu(args.gpu_mem, 0)
+    else:
+        config.disable_gpu()
+        config.set_cpu_math_library_num_threads(6)
+        if args.enable_mkldnn:
+            # cache 10 different shapes for mkldnn to avoid memory leak
+            config.set_mkldnn_cache_capacity(10)
+            config.enable_mkldnn()
+
+    # config.enable_memory_optim()
+    config.disable_glog_info()
+
+    if args.use_zero_copy_run:
+        config.delete_pass("conv_transpose_eltwiseadd_bn_fuse_pass")
+        config.switch_use_feed_fetch_ops(False)
+    else:
+        config.switch_use_feed_fetch_ops(True)
+
+    predictor = create_paddle_predictor(config)
+    input_names = predictor.get_input_names()
+    for name in input_names:
+        input_tensor = predictor.get_input_tensor(name)
+    output_names = predictor.get_output_names()
+    output_tensors = []
+    for output_name in output_names:
+        output_tensor = predictor.get_output_tensor(output_name)
+        output_tensors.append(output_tensor)
+    return predictor, input_tensor, output_tensors
+
+
 def draw_text_det_res(dt_boxes, img_path):
     src_im = cv2.imread(img_path)
     for box in dt_boxes: