Skip to content

[Automated PR (update-readme)]: Document TABPFN_MODEL_CACHE_SIZE and TABPFN_EXCLUDE_DEVICES env vars - #1244

Open
claude[bot] wants to merge 1 commit into
mainfrom
claude/gallant-hypatia-lgmpvo
Open

[Automated PR (update-readme)]: Document TABPFN_MODEL_CACHE_SIZE and TABPFN_EXCLUDE_DEVICES env vars#1244
claude[bot] wants to merge 1 commit into
mainfrom
claude/gallant-hypatia-lgmpvo

Conversation

@claude

@claude claude Bot commented Sep 7, 2026

Copy link
Copy Markdown
Contributor

⚠️ This PR was based on the routine defined in update-readme.
Routine owner: Philipp Singer — the person to ping if this PR misbehaves or the routine produces bad output.
As the automatically chosen reviewer, you are responsible for this PR.
Feel free to close or edit the PR, and either merge when ready or assign another reviewer if needed.
If the PR is not satisfactory, please edit the prompt in update-readme in the misc/claude-routines directory, or contact the routine owner.


What

The README's environment-variable FAQ ("What environment variables can I use to configure TabPFN?") lists the model-configuration variables a user can set, but two that exist in the codebase were missing. This adds them to the existing list.

Variable Source Notes
TABPFN_MODEL_CACHE_SIZE src/tabpfn/model_loading.py Opt-in LRU cache of built models; publicly announced in CHANGELOG 8.3.0 (#1162)
TABPFN_EXCLUDE_DEVICES src/tabpfn/utils.py Excludes device types from automatic (device="auto") device selection

Why

Weekly README-accuracy routine. The rest of the README was checked against the codebase and is consistent (Python versions, ModelVersion enum values, checkpoint filenames, CPU sample limits, and every other documented env var and default). These two omissions were the only discrepancy found. The change is scoped to two one-line additions in the existing list — no new sections or topics.

Verification

  • TABPFN_MODEL_CACHE_SIZE: read via os.environ.get("TABPFN_MODEL_CACHE_SIZE", "0") in _get_built_model_cache_size (model_loading.py); default 0 disables the cache.
  • TABPFN_EXCLUDE_DEVICES: read via os.getenv("TABPFN_EXCLUDE_DEVICES", "") in utils.py, split on commas and applied only in the device="auto" path.

🤖 Generated with Claude Code

https://claude.ai/code/session_01268Qq5aFytqL817RJxow2v


Generated by Claude Code

…ICES env vars

The environment-variable FAQ answers "What environment variables can I use
to configure TabPFN?" but omitted two configuration variables that exist in
the codebase:

- TABPFN_MODEL_CACHE_SIZE: opt-in LRU cache of built models
  (src/tabpfn/model_loading.py; announced in CHANGELOG 8.3.0)
- TABPFN_EXCLUDE_DEVICES: exclude device types from automatic device
  selection (src/tabpfn/utils.py)

Add both to the existing list to keep the reference accurate.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01268Qq5aFytqL817RJxow2v
@claude claude Bot added documentation no changelog needed PR does not require a changelog entry labels Sep 7, 2026
@claude
claude Bot requested a review from ggprior September 7, 2026 08:08
@LeoGrin
LeoGrin added this pull request to the merge queue Sep 11, 2026
@github-merge-queue
github-merge-queue Bot removed this pull request from the merge queue due to failed status checks Sep 11, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

documentation no changelog needed PR does not require a changelog entry

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants