noahho

Authors weren't properly listed on pypi and pepi with the emails

TabPFNRegressor fails on constant input data

1

### Describe the bug When fitting on constant regression target, errors ### Steps/Code to Reproduce ``` import numpy as np from tabpfn import TabPFNRegressor # Set random seed for reproducibility...

[Blocked] Use Scrub for data cleaning

4

Fix #138: NA handling in text columns Fix #163 Partially fixed by https://github.com/PriorLabs/TabPFN/pull/242 ## Summary - Add skrub>=0.3.0 dependency to handle mixed string/NA data - Integrate TableVectorizer in TabPFNClassifier to...

## Issue Description When running TabPFN consistency tests across different platforms (e.g., macOS vs Linux, x86 vs ARM), we've observed significant differences in model predictions. ## Current Observations: 1. Despite...

Improve Testing

1

### Describe the workflow you want to enable Better testing of the repository, especially to be able to accept PRs more confidently. ### Describe your proposed solution - Add more...

Document strategies to use more samples

4

Document dataset constraints for TabPFN more, document strategies to use for more samples (subsampling ensemble, SklearnBasedRandomForestTabPFN [tabpfn-extensions (https://github.com/PriorLabs/tabpfn-extensions/blob/dbc3f5da25821135602fdc4d95cc8c217afbc3b0/src/tabpfn_extensions/rf_pfn/SklearnBasedRandomForestTabPFN.py#L106])

Separate models into individual repositories on huggingface Standardize checkpoint naming (e.g., "model.ckpt") for consistency

Example usage and visualization and of full uncertainty mode: ``` reg = TabPFNRegressor() reg.fit(x, y_noisy) preds = reg.predict(x_test, output_type="full") fig, ax = plt.subplots(1, figsize=(12,6)) N = 10 #number of samples...

Add usage tips for TabPFN in README

1

Added important tips for using TabPFN effectively, including batch prediction mode, data preprocessing advice, GPU usage, and dataset size limitations. ## Issue Please link the corresponding GitHub issue. If an...