Pinning dependencies to known good versions. (#15)

soldni · web-flow · commit 92e17d1e96ff · 2025-08-12T15:44:35.000-07:00
* pinned to known versions

* changelog
diff --git a/.gitignore b/.gitignore
@@ -57,3 +57,6 @@ site/
 /wandb/
 /scratch/
 core
+
+# do not check in uv lock
+/uv.lock
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -7,6 +7,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## Unreleased
 
+- Pinned datasets, tokenizers, pyarrow versions.
+
 ## [v0.8.5](https://github.com/allenai/OLMo-in-loop-evals/releases/tag/v0.8.5) - 2025-07-20
 
 ### Removed
diff --git a/pyproject.toml b/pyproject.toml
@@ -15,12 +15,13 @@ license = { file = "LICENSE" }
 dependencies = [
     "torch",
     "torchmetrics",
-    "datasets",
-    "tokenizers",
+    "datasets>=3.6.0,<4",
     "cached-path",
     "requests",
     "packaging",
     "importlib_resources",
+    "tokenizers>=0.19.1,<0.20",
+    "pyarrow>=19.0,<20",    # datasets brings pyarrow, but version 21 breaks things
 ]
 
 [project.urls]
