TabPFN Common Utilities
Shared Python utilities used across the TabPFN ecosystem (the tabular foundation model).
Features
Data Processing Utilities
- Regression Results: Handling of prediction outputs with mean, median, mode, and quantiles
- Data Serialization: Convert between pandas DataFrames, NumPy arrays, and CSV formats
- Dataset Management: Load and preprocess standard ML datasets with proper train/test splits
- Preprocessing Configuration: Options for data transformation strategies
Cost Estimation
- Resource Planning: Estimation of computational costs and duration for TabPFN predictions
- Cloud Pricing: Useful for resource planning in cloud-based TabPFN services
- Task-Specific Calculations: Different cost models for classification vs regression tasks
Telemetry (optional, opt-out)
- Anonymous & Aggregated: No personal information or sensitive data is collected or transmitted
- Configurable: Can be disabled via environment variable
- Usage Patterns: Aggregate signals used to improve TabPFN
Installation
pip install tabpfn-common-utils
Or with uv:
uv add tabpfn-common-utils
Quick Start
Regression Results
from tabpfn_common_utils.regression_pred_result import RegressionPredictResult
# Handle regression prediction results
result = RegressionPredictResult({
"mean": [1.2, 2.3, 3.4],
"median": [1.1, 2.2, 3.3],
"mode": [1.0, 2.0, 3.0],
"quantile_0.25": [0.9, 1.9, 2.9],
"quantile_0.75": [1.5, 2.5, 3.5]
})
# Convert to basic representation for serialization
basic_repr = RegressionPredictResult.to_basic_representation(result)
Data Utilities
from tabpfn_common_utils.utils import get_example_dataset, serialize_to_csv_formatted_bytes
import pandas as pd
# Load example dataset
X_train, X_test, y_train, y_test = get_example_dataset("iris")
# Serialize data to CSV bytes
csv_bytes = serialize_to_csv_formatted_bytes(X_train)
Telemetry
from tabpfn_common_utils.telemetry import ProductTelemetry
# Initialize telemetry service (anonymous; opt-out)
telemetry = ProductTelemetry()
# Track usage events
telemetry.capture(...)
# Disable via environment variable
export TABPFN_DISABLE_TELEMETRY=1
Telemetry notes
- Anonymous and aggregated only — no user identification or tracking
- Disabled by setting
TABPFN_DISABLE_TELEMETRY=1 - Open source — see
src/tabpfn_common_utils/telemetry/for what is sent
Development
Setup
# Install dependencies
uv sync
# Activate virtual environment
source .venv/bin/activate
# Run tests
uv run pytest
# Type checking
uv run pyright
# Code formatting
uv run ruff check --fix
Adding Dependencies
# Add runtime dependency
uv add <package_name>
# Add development dependency
uv add --group dev <package_name>
License
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.
Contributing
Contributions are welcome! Please ensure all code passes type checking and formatting requirements.
Links
4 maintainers
Metadata
4 maintainers
Release files for tabpfn-common-utils 0.2.23
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| tabpfn_common_utils-0.2.23.tar.gz | 2.0 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| tabpfn_common_utils-0.2.23-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 2.0 MB
Release files / tabpfn_common_utils-0.2.23.tar.gz
| Download URL | tabpfn_common_utils-0.2.23.tar.gz |
|---|---|
| Size | 2.0 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
4267831070efbcf1baae77e48e28a78a11e87ecda5dcb60744f81c69e7163258
|
|
BLAKE2b-256 checksum How to use checksums |
ba8cd1d8e744ade10490d56f653bb0cdba180ff8726c78202a66302ab2578b58
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jun 22, 2026.
Transparency logRelease files / tabpfn_common_utils-0.2.23-py3-none-any.whl
| Download URL | tabpfn_common_utils-0.2.23-py3-none-any.whl |
|---|---|
| Size | 40.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
e456e917cf2fbe44c3a44352c737ed3194ff572f2ee5141294a59dae053b8af2
|
|
BLAKE2b-256 checksum How to use checksums |
4f5f03498070cb10d3998888395cee67016cc6e937138a2e89d9a44ae177dbd7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jun 22, 2026.
Transparency log
