pytorch · vfdev-5 · Aug 28, 2020 · Aug 27, 2020 · Aug 27, 2020 · Aug 27, 2020
diff --git a/README.md b/README.md
@@ -296,7 +296,26 @@ version as dependency):
 ```bash
 conda install ignite -c pytorch-nightly
 ```
+## Docker Images
 
+### Using pre-built images
+
+Pull a pre-built docker image from [our Docker Hub](https://hub.docker.com/u/pytorchignite) and run it with docker v19.03+.
+
+```bash
+docker run --gpus all -it -v $PWD:/workspace/project --network=host --shm-size 16G pytorchignite/base:latest
+```
+
+Available pre-built images are :
+
+- `pytorchignite/base:latest`  
+- `pytorchignite/apex:latest`
+- `pytorchignite/vision:latest` 
+- `pytorchignite/apex-vision:latest`
+- `pytorchignite/nlp:latest`
+- `pytorchignite/apex-nlp:latest`
+
+For more details, see [here](docker).
 
 <!-- ############################################################################################################### -->
 

diff --git a/docker/README.md b/docker/README.md
@@ -20,5 +20,14 @@ We provide Dockerfiles in order to build containerized execution environment tha
 ## How to use
 
 ```bash
-docker run --rm -it -v $PWD:/workspace/project --network=host --shm-size 16G pytorchignite/base:latest
+docker run -it -v $PWD:/workspace/project --network=host --shm-size 16G pytorchignite/base:latest
+```
+
+## Building the image yourself
+
+Dockerfiles are supplied to build images with dependencies that ease the use of Ignite for computer vision / NLP tasks:
+
+```bash
+cd main
+docker build -t pytorchignite/base:latest -f Dockerfile.base . 
 ```
diff --git a/setup.cfg b/setup.cfg
@@ -10,7 +10,7 @@ ignore = E402, E721
 max_line_length = 120
 
 [isort]
-known_third_party=dill,matplotlib,numpy,pytest,setuptools,sklearn,torch,torchvision,trains
+known_third_party=dill,matplotlib,numpy,pkg_resources,pytest,setuptools,sklearn,torch,torchvision,trains
 multi_line_output=3
 include_trailing_comma=True
 force_grid_wrap=0