Merge branch 'dygraph' of https://github.com/PaddlePaddle/PaddleOCR into multi_languages

paddleocr.py

@@ -290,7 +290,9 @@ class PaddleOCR(predict_system.TextSystem):
             image_file = img
             img, flag = check_and_read_gif(image_file)
             if not flag:
-                img = cv2.imread(image_file)
+                with open(image_file, 'rb') as f:
+                    np_arr = np.frombuffer(f.read(), dtype=np.uint8)
+                    img = cv2.imdecode(np_arr, cv2.IMREAD_COLOR)
             if img is None:
                 logger.error("error in loading image:{}".format(image_file))
                 return None

ppocr/data/__init__.py

@@ -51,7 +51,7 @@ signal.signal(signal.SIGINT, term_mp)
 signal.signal(signal.SIGTERM, term_mp)
 
 
-def build_dataloader(config, mode, device, logger):
+def build_dataloader(config, mode, device, logger, seed=None):
     config = copy.deepcopy(config)
 
     support_dict = ['SimpleDataSet', 'LMDBDateSet']
@@ -61,7 +61,7 @@ def build_dataloader(config, mode, device, logger):
     assert mode in ['Train', 'Eval', 'Test'
                     ], "Mode should be Train, Eval or Test."
 
-    dataset = eval(module_name)(config, mode, logger)
+    dataset = eval(module_name)(config, mode, logger, seed)
     loader_config = config[mode]['loader']
     batch_size = loader_config['batch_size_per_card']
     drop_last = loader_config['drop_last']

ppocr/data/lmdb_dataset.py

@@ -21,7 +21,7 @@ from .imaug import transform, create_operators
 
 
 class LMDBDateSet(Dataset):
-    def __init__(self, config, mode, logger):
+    def __init__(self, config, mode, logger, seed=None):
         super(LMDBDateSet, self).__init__()
 
         global_config = config['Global']