diff --git a/PaddleCV/PaddleDetection/slim/quantization/README.md b/PaddleCV/PaddleDetection/slim/quantization/README.md
index c151678b7c08f8eb6fa5b9bf516832e6f75b3dea..d22f636cc60f3c3cbe568b6a3a4bb3d7236dacf6 100644
--- a/PaddleCV/PaddleDetection/slim/quantization/README.md
+++ b/PaddleCV/PaddleDetection/slim/quantization/README.md
@@ -53,7 +53,7 @@ python compress.py \
     -d "../../dataset/voc" \
     -o max_iters=258 \
     LearningRate.base_lr=0.0001 \
-    LearningRate.schedulers='[!PiecewiseDecay {gamma: 0.1, milestones: [258, 516]}]' \
+    LearningRate.schedulers="[!PiecewiseDecay {gamma: 0.1, milestones: [258, 516]}]" \
     pretrain_weights=https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1_voc.tar \
     YoloTrainFeed.batch_size=64
 ```
@@ -78,7 +78,7 @@ python compress.py \
     -d "../../dataset/voc" \
     -o max_iters=258 \
     LearningRate.base_lr=0.0001 \
-    LearningRate.schedulers='[!PiecewiseDecay {gamma: 0.1, milestones: [258, 516]}]' \
+    LearningRate.schedulers="[!PiecewiseDecay {gamma: 0.1, milestones: [258, 516]}]" \
     pretrain_weights=https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1_voc.tar \
     YoloTrainFeed.batch_size=64
   
@@ -93,7 +93,7 @@ python compress.py \
     -d "../../dataset/voc" \
     -o max_iters=516 \
     LearningRate.base_lr=0.00005 \
-    LearningRate.schedulers='[!PiecewiseDecay {gamma: 0.1, milestones: [516, 1012]}]' \
+    LearningRate.schedulers="[!PiecewiseDecay {gamma: 0.1, milestones: [516, 1012]}]" \
     pretrain_weights=https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1_voc.tar \
     YoloTrainFeed.batch_size=32
 ```