diff --git a/README.md b/README.md
index 409473a..75a7712 100644
--- a/README.md
+++ b/README.md
@@ -31,9 +31,13 @@ We will also explore more advanced models such as graph neural networks (GNNs),
 
 ## Setup
 
-### Vanderbilt Jupyterhub (Recommended!)
+### Purdue Analysis Facility (New and recommended!)
 
-The recommended method for running the tutorials live is the Vanderbilt Jupyterhub, follow the instructions [here](https://fnallpc.github.io/machine-learning-hats/setup/vanderbilt-jupyterhub/vanderbilt.html).
+The recommended method for running the tutorials live is the Purdue AF, follow the instructions [here](https://fnallpc.github.io/machine-learning-hats/setup/purdue/purdue.html).
+
+### Vanderbilt Jupyterhub
+
+Another option is the Vanderbilt Jupyterhub, instructions [here](https://fnallpc.github.io/machine-learning-hats/setup/vanderbilt-jupyterhub/vanderbilt.html).
 
 ### FNAL LPC
 
diff --git a/machine-learning-hats/_toc.yml b/machine-learning-hats/_toc.yml
index ea41e8a..b170b66 100644
--- a/machine-learning-hats/_toc.yml
+++ b/machine-learning-hats/_toc.yml
@@ -6,6 +6,7 @@ root: index
 parts:
   - caption: Setup
     chapters:
+    - file: setup/purdue/purdue
     - file: setup/vanderbilt-jupyterhub/vanderbilt
     - file: setup/lpc
     - file: setup-libraries
diff --git a/machine-learning-hats/notebooks/1-datasets-uproot.ipynb b/machine-learning-hats/notebooks/1-datasets-uproot.ipynb
index 2d3f394..99514f1 100644
--- a/machine-learning-hats/notebooks/1-datasets-uproot.ipynb
+++ b/machine-learning-hats/notebooks/1-datasets-uproot.ipynb
@@ -117,7 +117,7 @@
    },
    "source": [
     "## Convert tree to `pandas` DataFrames\n",
-    "In my opinion, `pandas` DataFrames are a more convenient/flexible data container in python: https://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.html. "
+    "`pandas` DataFrames can be a more convenient/flexible data container in python, especially for large amounts of data: https://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.html. "
    ]
   },
   {
diff --git a/machine-learning-hats/setup/purdue/folders.png b/machine-learning-hats/setup/purdue/folders.png
new file mode 100644
index 0000000..3760222
Binary files /dev/null and b/machine-learning-hats/setup/purdue/folders.png differ
diff --git a/machine-learning-hats/setup/purdue/git.png b/machine-learning-hats/setup/purdue/git.png
new file mode 100644
index 0000000..1e97976
Binary files /dev/null and b/machine-learning-hats/setup/purdue/git.png differ
diff --git a/machine-learning-hats/setup/purdue/purdue.md b/machine-learning-hats/setup/purdue/purdue.md
new file mode 100644
index 0000000..e180761
--- /dev/null
+++ b/machine-learning-hats/setup/purdue/purdue.md
@@ -0,0 +1,33 @@
+# Purdue Analysis Facility
+
+## 1. Sign-in
+
+See the [Getting Started](https://analysis-facility.physics.purdue.edu/en/latest/doc-getting-started.html) guide and the rest of the documentation for details.
+
+Point your browser to https://cms.geddes.rcac.purdue.edu/hub and log in with your CERN or FNAL account.
+
+Create an instance with the default resources.
+
+
+## 2. Clone this repository
+
+1. Once the session starts, open the Git sidebar menu:
+
+![git menu](git.png)
+
+2. Click "Clone a Repository". 
+
+3. Copy and paste the repository URL: https://github.com/FNALLPC/machine-learning-hats.git and Clone with the default options.
+
+4. You should now see the `machine-learning-hats` directory in your file browser:
+
+![folders](folders.png)
+
+Open it and navigate to `machine-learning-hats` -> `notebooks` 
+
+
+## 3. Notebooks
+
+Open up a notebook and use the `Python3 kernel (default)` kernel.  You can now the run the notebook by pressing `Shift + Enter`, one cell at a time.
+
+