Multiple updates

Author: nomisto

README.md

@@ -5,4 +5,8 @@ https://www.moncefbelyamani.com/making-github-pages-work-with-latest-jekyll/
 
 ## Generate doc for loader and evaluator
 
-from src folder run ```pdoc --html openbiolink/obl2021/obl2021.py --force```
+from src folder run ```pdoc --html openbiolink/obl2021/obl2021.py --force```
+
+## Local test
+
+```bundle exec jekyll serve --watch```

_sass/_page.scss

@@ -13,13 +13,24 @@
     @include flexbox{
         flex-direction: column;
         justify-content: flex-start;
-        align-items: center;
+        align-items: flex-start;
     }
 
     img{
         max-width: 100%;
     }
 
+    figure{
+        margin-left: auto;
+        margin-right: auto;
+        margin-top: 0px;
+        margin-bottom: 0px;
+    }
+
+    table{
+        margin-left: auto;
+        margin-right: auto;
+    }
 
     h1{
         margin-right:auto;

challenge.md

@@ -12,8 +12,35 @@ To foster progress in link prediction on large-scale, heterogeneous biomedical d
 
 # Timeline
 
-| Challenge start           |   |   |
-| Submission deadline       | 01.12.2021 - 23:59 GMT-0  |   |
+
+| Task                      |     Deadline              |               |
+|---------------------------|---------------------------|---------------|
+| Team registration         |            -              |  [Link](TBD)  |
+| Submission deadline       | 01.12.2021 - 23:59 GMT-0  |  [Link](TBD)  |
+
+There is no deadline for registering your team however, you'll need to register to OpenBioLink2021 before being able to make a submission.
+
+# Prizes
+
++ Best predictive performance prize (**750 €**)
++ Best explainability prize (**250 €**)
+
+Best predictive performance prize can only be given to submissions that improve on current baseline results. Best explainability prize will be awarded based on evaluation by the OpenBioLink Challenge team. Furthermore, results of the challenge will be published in a joint paper on arXiv (and potentially a peer-reviewed journal). Challenge participants are of course free to also publish about their work in their own scientific papers.
+
+# Baseline results
+
+The code of our embedding baseline models is located [here](https://github.com/nomisto/openbiolink-2021-embedding-baseline),  the code for our explainable baseline is located [here](https://github.com/nomisto/openbiolink-2021-symbolic-baseline).
+
+| Model          | Hits@10 | Explainable? |
+|----------------|---------|--------------|
+| AnyBURL        | 0.000   | X            |
+| RESCAL         | 0.000   |              |
+| TransR         | 0.000   |              |
+| DistMult       | 0.000   |              |
+| ComplEx        | 0.000   |              |
+| RotatE         | 0.000   |              |
+| TransE         | 0.000   |              |
+
 
 # Questions
 

dataset.md

@@ -12,7 +12,7 @@ The OpenBioLink2021 Dataset is a highly challenging benchmark dataset containing
 |-----------|---------|---------|------------|-------------|
 | 4,192,002 | 186,301 | 180,964 | 180,992    | 28          |
 
-The dataset can be loaded with the provided DataLoader, which is further documented [here](../obl2021.html#obl2021.OBL2021Dataset)
+The dataset can be downloaded [here](https://zenodo.org/record/5361324/files/KGID_HQ_DIR.zip?download=1) or loaded with the provided DataLoader, which is further documented [here](../obl2021.html#obl2021.OBL2021Dataset)
 
 {% highlight python %}