diff --git a/experiments/edbo_demo_and_simulations.ipynb b/experiments/edbo_demo_and_simulations.ipynb
index fe63a8d..f2aa5e4 100644
--- a/experiments/edbo_demo_and_simulations.ipynb
+++ b/experiments/edbo_demo_and_simulations.ipynb
@@ -15,7 +15,11 @@
    "source": [
     "## Software\n",
     "\n",
-    "Experimental Design via Bayesian Optimization: *edbo* is a practical python implementation of Bayesian optimization. The current build has 16 modules with essential and experimental features. Documentation for each module can be found in the source code or by using help(edbo.module). However, when the GitHub site goes live *edbo* will have a documentation page. The main Bayesian optimization program is accessed through the edbo.bro module. The main BO class, edbo.bro.BO, is a full featured program for choosing initial experiments, running BO on human-in-the-loop or computational objectives, modeling, and analyzing results. We used this class to run experiments for the paper. Note: edbo.bro.BO parameters are preset to those optimized for DFT encodings in the paper. See supplementary information for parameters optimized for one-hot-encoded and Mordred encoded reaction spaces.\n",
+    "Experimental Design via Bayesian Optimization: *edbo* is a practical python implementation of Bayesian optimization. The current build has 16 modules with essential and experimental features. Documentation for the main modules can be found on the documentation page or by using help(edbo.module).\n",
+    "\n",
+    "**Documentation:** https://b-shields.github.io/edbo/index.html\n",
+    "\n",
+    "The main Bayesian optimization program is accessed through the edbo.bro module. The main BO class, edbo.bro.BO, is a full featured program for choosing initial experiments, running BO on human-in-the-loop or computational objectives, modeling, and analyzing results. We used this class to run experiments for the paper. Note: edbo.bro.BO parameters are preset to those optimized for DFT encodings in the paper. See supplementary information for parameters optimized for one-hot-encoded and Mordred encoded reaction spaces.\n",
     "\n",
     "It can be fun to play around with the software in optimizing some 1D objectives. This is a nice way to get to know some of the features and be able to visualize the results. Below is a brief demonstration of an arbitrary 1D objective with a discretized domain. While edbo is designed to work with human-in-the-loop experimentation you can also use computational objectives (we also demonstrate this feature below). For example, this is how we accomplished self-optimization of the GP hyperparameters using *edbo*.\n",
     "\n",