Add Pytorch and set sklearn version in Dockerfile

deployment/docker/Dockerfile

@@ -70,8 +70,16 @@ RUN python3 -m pip --no-cache-dir install torch torchvision
 #
 RUN python3 -m pip --no-cache-dir install Keras==2.1.6
 
-#sklearn
-RUN python3 -m pip --no-cache-dir install scikit-learn
+#
+#PyTorch 
+#
+RUN python3 -m pip --no-cache-dir install torch==0.4.1
+RUN python3 -m pip install torchvision==0.2.1
+
+#
+#sklearn 0.20.0
+#
+RUN python3 -m pip --no-cache-dir install scikit-learn==0.20.0
 
 ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/root/.local/bin:/usr/bin:/bin:/sbin
 

deployment/docker/README.md

@@ -9,6 +9,8 @@ numpy 1.14.3,scipy 1.1.0
 TensorFlow 1.5.0
 PyTorch (Stable)
 Keras 2.1.6
+PyTorch 0.4.1
+scikit-learn 0.20.0
 NNI v0.3
 ```
 You can take this Dockerfile as a reference for your own customized Dockerfile. 

tools/nni_cmd/nnictl.py

@@ -73,7 +73,7 @@ def parse_args():
 
     #parse stop command
     parser_stop = subparsers.add_parser('stop', help='stop the experiment')
-    parser_stop.add_argument('id', nargs='?', help='the id of experiment, use "all" for all experiments')
+    parser_stop.add_argument('id', nargs='?', help='the id of experiment, use \'all\' to stop all running experiments')
     parser_stop.set_defaults(func=stop_experiment)
 
     #parse trial command