diff --git a/ppstructure/layout/README.md b/ppstructure/layout/README.md
index 0931702a7cf411e6589a1375e014a7374442f9f0..3a4f5291763e34c8aec2c5b327d40a459bb4be1e 100644
--- a/ppstructure/layout/README.md
+++ b/ppstructure/layout/README.md
@@ -52,7 +52,7 @@ The following figure shows the result, with different colored detection boxes re
 |   threshold    |              threshold of prediction score               |     0.5     |                              \                               |
 |  input_shape   |                 picture size of reshape                  | [3,640,640] |                              \                               |
 |   batch_size   |                    testing batch size                    |      1      |                              \                               |
-|   label_map    |                  category mapping table                  |    None     | Setting config_ path, it can be none, and the label is automatically obtained according to the dataset name_ map |
+|   label_map    |                  category mapping table                  |    None     | Setting config_ path, it can be none, and the label is automatically obtained according to the dataset name_ map, You need to specify it manually when setting model_path |
 |  enforce_cpu   |                    whether to use CPU                    |    False    |      False to use GPU, and True to force the use of CPU      |
 | enforce_mkldnn | whether mkldnn acceleration is enabled in CPU prediction |    True     |                              \                               |
 |   thread_num   |                the number of CPU threads                 |     10      |                              \                               |
diff --git a/ppstructure/layout/README_ch.md b/ppstructure/layout/README_ch.md
index 6fec748b7683264f5b4a7d29c0e51c84773425ba..825ff62b116171fda277528017292434bd75b941 100644
--- a/ppstructure/layout/README_ch.md
+++ b/ppstructure/layout/README_ch.md
@@ -52,7 +52,7 @@ show_img.show()
 |   threshold    |       预测得分的阈值        |     0.5     |                              \                               |
 |  input_shape   |     reshape之后图片尺寸     | [3,640,640] |                              \                               |
 |   batch_size   |       测试batch size        |      1      |                              \                               |
-|   label_map    |         类别映射表          |    None     | 设置config_path时，可以为None，根据数据集名称自动获取label_map |
+|   label_map    |         类别映射表          |    None     | 设置config_path时，可以为None，根据数据集名称自动获取label_map,设置model_path时需要手动指定 |
 |  enforce_cpu   |     代码是否使用CPU运行     |    False    |         设置为False表示使用GPU，True表示强制使用CPU          |
 | enforce_mkldnn | CPU预测中是否开启MKLDNN加速 |    True     |                              \                               |
 |   thread_num   |        设置CPU线程数        |     10      |                              \                               |
diff --git a/ppstructure/predict_system.py b/ppstructure/predict_system.py
index 3f3dc65875a20b3f66403afecfd60f04e3d83d61..3ae52fdd703670c4250f1b4a440004fa8b9082ad 100644
--- a/ppstructure/predict_system.py
+++ b/ppstructure/predict_system.py
@@ -58,6 +58,7 @@ class OCRSystem(object):
             self.table_layout = lp.PaddleDetectionLayoutModel(
                 config_path=config_path,
                 model_path=model_path,
+                label_map=args.layout_label_map,
                 threshold=0.5,
                 enable_mkldnn=args.enable_mkldnn,
                 enforce_cpu=not args.use_gpu,
diff --git a/ppstructure/utility.py b/ppstructure/utility.py
index ce7a801b1bb4094d3f4d2ba467332c6763ad6287..43cb0b0873812baf3ce2dc689fb62f1d0ca2c551 100644
--- a/ppstructure/utility.py
+++ b/ppstructure/utility.py
@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import ast
 from PIL import Image
 import numpy as np
 from tools.infer.utility import draw_ocr_box_txt, init_args as infer_args
@@ -34,7 +35,11 @@ def init_args():
         "--layout_path_model",
         type=str,
         default="lp://PubLayNet/ppyolov2_r50vd_dcn_365e_publaynet/config")
-
+    parser.add_argument(
+        "--layout_label_map",
+        type=ast.literal_eval,
+        default=None,
+        help='label map according to ppstructure/layout/README_ch.md')
     # params for ser
     parser.add_argument("--model_name_or_path", type=str)
     parser.add_argument("--max_seq_length", type=int, default=512)