diff --git a/configs/fast_scnn_pet.yaml b/configs/fast_scnn_pet.yaml
index 02fdef8ea9a15c3c132d80b69cf3e4f6d0876c1f..2b9b659f18735324eeb282185a082e5c52e2a063 100644
--- a/configs/fast_scnn_pet.yaml
+++ b/configs/fast_scnn_pet.yaml
@@ -31,7 +31,7 @@ MODEL:
     DEFAULT_NORM_TYPE: "bn"
 
 TRAIN:
-    PRETRAINED_MODEL_DIR: "./pretrained_model/fast_scnn_cityscape/"
+    PRETRAINED_MODEL_DIR: "./pretrained_model/fast_scnn_cityscapes/"
     MODEL_SAVE_DIR: "./saved_model/fast_scnn_pet/"
     SNAPSHOT_EPOCH: 10
 TEST:
diff --git a/turtorial/finetune_fast_scnn.md b/turtorial/finetune_fast_scnn.md
index 0ad8bde68bf8783e72f92d4e8e34482c55c32065..0397031632fbb23cd433842e9ae88b7953e4d6f6 100644
--- a/turtorial/finetune_fast_scnn.md
+++ b/turtorial/finetune_fast_scnn.md
@@ -68,7 +68,7 @@ AUG:
     FIX_RESIZE_SIZE: (512, 512)
 BATCH_SIZE: 4
 TRAIN:
-    PRETRAINED_MODEL_DIR: "./pretrained_model/fast_scnn_cityscape/"
+    PRETRAINED_MODEL_DIR: "./pretrained_model/fast_scnn_cityscapes/"
     MODEL_SAVE_DIR: "./saved_model/fast_scnn_pet/"
     SNAPSHOT_EPOCH: 10
 TEST:
@@ -114,6 +114,6 @@ python pdseg/eval.py --use_gpu --cfg ./configs/fast_scnn_pet.yaml
 | ICNet/bn |(1024, 2048) |8.76ms| 0.6831 |
 | Fast-SCNN/bn | (1024, 2048) |6.28ms| 0.6964 |
 
-上述测试环境为v100.
+上述测试环境为v100. 测试使用paddle的推理接口[zero_copy]()的方式，模型输出是类别，即argmax后的类别。