💾📉 Improve Memory-Efficiency of Converting OGB Datasets to TriplesFac…

…tory (pykeen#1253) - update loading of OGB datasets to be more memory-efficient - improve typing in OGB datasets module - fix OGB BioKG entity count - misc: enforce encoding when reading/writing README Fix pykeen#1252
nicolafan · Apr 15, 2023 · 3f8d0e2 · 3f8d0e2
1 parent fdce5b1
commit 3f8d0e2
Show file tree

Hide file tree

Showing 4 changed files with 249 additions and 76 deletions.
diff --git a/README.md b/README.md
@@ -141,7 +141,7 @@ have a suggestion for another dataset to include in PyKEEN, please let us know
 | Kinships                           | [`pykeen.datasets.Kinships`](https://pykeen.readthedocs.io/en/latest/api/pykeen.datasets.Kinships.html)             | [Kemp *et al*., 2006](https://www.aaai.org/Papers/AAAI/2006/AAAI06-061.pdf)                                             |        104 |          25 |     10686 |
 | Nations                            | [`pykeen.datasets.Nations`](https://pykeen.readthedocs.io/en/latest/api/pykeen.datasets.Nations.html)               | [`ZhenfengLei/KGDatasets`](https://github.com/ZhenfengLei/KGDatasets)                                                   |         14 |          55 |      1992 |
 | NationsL                           | [`pykeen.datasets.NationsLiteral`](https://pykeen.readthedocs.io/en/latest/api/pykeen.datasets.NationsLiteral.html) | [`pykeen/pykeen`](https://github.com/pykeen/pykeen)                                                                     |         14 |          55 |      1992 |
-| OGB BioKG                          | [`pykeen.datasets.OGBBioKG`](https://pykeen.readthedocs.io/en/latest/api/pykeen.datasets.OGBBioKG.html)             | [Hu *et al*., 2020](https://arxiv.org/abs/2005.00687)                                                                   |      45085 |          51 |   5088433 |
+| OGB BioKG                          | [`pykeen.datasets.OGBBioKG`](https://pykeen.readthedocs.io/en/latest/api/pykeen.datasets.OGBBioKG.html)             | [Hu *et al*., 2020](https://arxiv.org/abs/2005.00687)                                                                   |      93773 |          51 |   5088434 |
 | OGB WikiKG2                        | [`pykeen.datasets.OGBWikiKG2`](https://pykeen.readthedocs.io/en/latest/api/pykeen.datasets.OGBWikiKG2.html)         | [Hu *et al*., 2020](https://arxiv.org/abs/2005.00687)                                                                   |    2500604 |         535 |  17137181 |
 | OpenBioLink                        | [`pykeen.datasets.OpenBioLink`](https://pykeen.readthedocs.io/en/latest/api/pykeen.datasets.OpenBioLink.html)       | [Breit *et al*., 2020](https://doi.org/10.1093/bioinformatics/btaa274)                                                  |     180992 |          28 |   4563407 |
 | OpenBioLink LQ                     | [`pykeen.datasets.OpenBioLinkLQ`](https://pykeen.readthedocs.io/en/latest/api/pykeen.datasets.OpenBioLinkLQ.html)   | [Breit *et al*., 2020](https://doi.org/10.1093/bioinformatics/btaa274)                                                  |     480876 |          32 |  27320889 |

diff --git a/src/pykeen/cli.py b/src/pykeen/cli.py
@@ -601,7 +601,7 @@ def readme(check: bool):
     new_readme = get_readme()
 
     if check:
-        with open(readme_path) as file:
+        with open(readme_path, encoding="utf8") as file:
             old_readme = file.read()
         if new_readme.strip() != old_readme.strip():
             click.secho(
@@ -616,7 +616,7 @@ def readme(check: bool):
 
             sys.exit(-1)
 
-    with open(readme_path, "w") as file:
+    with open(readme_path, "w", encoding="utf8") as file:
         print(new_readme, file=file)  # noqa:T201