diff --git a/rcnn/MaskRcnnInferencePlugin.h b/rcnn/MaskRcnnInferencePlugin.h
index bbf591c..ddf97ff 100644
--- a/rcnn/MaskRcnnInferencePlugin.h
+++ b/rcnn/MaskRcnnInferencePlugin.h
@@ -25,7 +25,7 @@ int maskRcnnInference(int batchSize,
 class MaskRcnnInferencePlugin : public IPluginV2Ext {
     int _detections_per_im;
     int _output_size;
-    int _num_classes;
+    int _num_classes = 1;
 
  protected:
     void deserialize(void const* data, size_t length) {
diff --git a/rcnn/README.md b/rcnn/README.md
index 79aab10..ab94d5a 100644
--- a/rcnn/README.md
+++ b/rcnn/README.md
@@ -124,7 +124,7 @@ sudo ./rcnn -d faster.engine ../samples
 
 - you can build fasterRcnn with maskRcnn weights file.
 
-- do initializing for _pre_nms_topk in RpnNmsPlugin and _count in BatchedNmsPlugin inside class to prevent error assert, because the configurePlugin function is implemented after clone() and before serialize(). one can also set it through constructor.
+- do initializing for _pre_nms_topk in RpnNmsPlugin,  _count in BatchedNmsPlugin and _num_classes in MaskRcnnInferencePlugin inside class to prevent error assert, because the configurePlugin function is implemented after clone() and before serialize(). one can also set it through constructor.
 
 ## Quantization