diff --git a/docs/usage/convert.md b/docs/usage/convert.md
index 28ed32064..b6701a369 100644
--- a/docs/usage/convert.md
+++ b/docs/usage/convert.md
@@ -1,7 +1,7 @@
 # Convert your training result to FPGA ready format
 
 ```
-$ PYTHONPATH=.:lmnet:dlk/python/dlk python blueoil/cmd/main.py convert -e test_20180101000000
+$ python blueoil/cmd/main.py convert -e test_20180101000000
 
 Usage:
   main.py convert [OPTIONS]
diff --git a/docs/usage/init.md b/docs/usage/init.md
index 32fabd6b5..40f6ab73d 100644
--- a/docs/usage/init.md
+++ b/docs/usage/init.md
@@ -2,7 +2,7 @@
 
 You can generate your configuration file interactively by running `python blueoil/cmd/main.py init`.
 
-    $ PYTHONPATH=.:lmnet:dlk/python/dlk python blueoil/cmd/main.py init
+    $ python blueoil/cmd/main.py init
 
 `python blueoil/cmd/main.py init` generates a configuration file used to train your new model.
 
diff --git a/docs/usage/train.md b/docs/usage/train.md
index 4d5e88079..87e0c5286 100644
--- a/docs/usage/train.md
+++ b/docs/usage/train.md
@@ -2,7 +2,7 @@
 
 
 ```
-$ PYTHONPATH=.:lmnet:dlk/python/dlk python blueoil/cmd/main.py train -c config/test.yml
+$ python blueoil/cmd/main.py train -c config/test.py
 
 Usage:
   main.py train [OPTIONS]