Updated yaml

rescience_report/metadata-draft.yaml

@@ -5,8 +5,7 @@
 #  - For a successful replication, it shoudl be prefixed with "[Re]"
 #  - For a failed replication, it should be prefixed with "[¬Re]"
 #  - For other article types, no instruction (but please, not too long)
-title: "Learning Neural PDE Solvers with Convergence Guarantees
-        ICLR Reproducibility Challenge 2019"
+title: "Learning Neural PDE Solvers with Convergence Guarantees ICLR Reproducibility Challenge 2019"
 
 # List of authors with name, orcid number, email and affiliation
 # Affiliation "*" means contact author
@@ -57,7 +56,8 @@ replication:
  - doi:  # Regular digital object identifier
 
 # Don't forget to surround abstract with double quotes
-   abstract: # Bibliography file (yours)
+   abstract: 'This paper is part of the ICLR Reproducibility Challenge 2019. We tried to replicate the results and algorithm for training an iterative partial differential equation solver by interpreting        the solver as a linear convolutional neural network and optimizing the weights of the convolutional kernels. We can replicate the results of the original paper \\cite{original_paper}\\footnote{We         based our this report on the version from the 10 October 2018. We noted no differences to the newest version from the 23 November 2018 with regard to this report.}, obtaining a general solver, which      generalizes well to a wide variety of geometries and boundary conditions, while achieving high speed ups compared to the baseline solver and guaranteeing convergence.\\footnote{The code, written in       Python using PyTorch, can be found on Github: \\url{https://github.com/francescobardi/pde_solver_deep_learned'
+
 bibliography: iclr2019_conference.bib
 
 # Type of the article