Added sources for QM9 dataset

report/main.tex

@@ -67,7 +67,7 @@ \section{Introduction}
 To verify and support our theoretical foundations, we also run benchmarks against other approaches used for learning on graph data~\cite{horn_topological_2022}
 
 \section{Data}
-We will be using the QM9 dataset[LINK TO DATASET?] as real-world data to benchmark our models against
+We will be using the QM9 dataset~\cite{blum}\cite{rupp} as real-world data to benchmark our models against
 other models from the cited papers and to test the metrics we will develop.
 
 We will also be creating and using synthetic datasets as this will give us more control over the data and allow us to create specific topological features and properties that we can then use to test the metrics and models.
@@ -78,9 +78,6 @@ \section{Methods}
 	\item Docker will be used to containerise our experiments and replications to ensure reproducibility as well as make it easier to run experiments on different machines.
 \end{itemize}
 
-
-
-[ADD MORE STUFF]
 \section{Learning Objectives}
 
 

report/sources.bib

@@ -234,4 +234,24 @@ @misc{li_deeper_2018
 	year = {2018},
 	note = {arXiv:1801.07606 [cs, stat]},
 	keywords = {Computer Science - Machine Learning, Statistics - Machine Learning},
-}
+}
+
+% Sources for QM9
+@article{blum,
+  author  = {L. C. Blum and J.-L. Reymond},
+  title   = {970 Million Druglike Small Molecules for Virtual Screening in the Chemical Universe Database {GDB-13}},
+  journal = "J. Am. Chem. Soc.",
+  volume  = 131,
+  pages   = 8732,
+  year    = 2009
+}
+
+@ARTICLE{rupp,
+  author  = {M. Rupp and A. Tkatchenko and K.-R. M\"uller and O. A. von Lilienfeld},
+  title   = {Fast and accurate modeling of molecular atomization energies with machine learning},
+  journal = "Physical Review Letters",
+  year    = 2012,
+  volume  = 108,
+  pages   = 058301
+}
+