skills.rest
dianaprior
dianapriorCOMMUNITY
Explore all skills by this author →
GitHub

tabpfn-explore

Profile tabular competition datasets and validate train-test distribution shift.

5|1|Updated Feb 25, 2026
Search Tags:#cross-validation#eda#data-profiling#kaggle#adversarial-validation#tabpfn#api-budget
One-click install
npx skills add https://github.com/dianaprior/kaggle-competition-agent-skill --skill tabpfn-explore
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tabpfn-explore
Source: https://github.com/dianaprior/kaggle-competition-agent-skill/tree/main/.claude/skills/tabpfn-explore
Command: npx skills add https://github.com/dianaprior/kaggle-competition-agent-skill --skill tabpfn-explore

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enables rapid, systematic exploration and validation of tabular competition datasets so you can detect leakage, distribution shift, and prepare consistent cross-validation splits before any modeling or API calls.

Core Features & Use Cases

  • Competition reconnaissance: document evaluation metric, task type, dataset size, and known pitfalls to guide modeling choices.
  • Exploratory Data Analysis: profile missingness, class balance, high-cardinality categoricals, duplicates, and high correlations that affect model design.
  • Adversarial validation: run train-vs-test classifiers to quantify distribution shift and surface the features driving it.
  • CV scheme and budget checks: define and save reproducible folds (StratifiedKFold, GroupKFold, TimeSeriesSplit) and verify TabPFN API cell budget constraints.
  • Deliverables: cleaned X_train/X_test/y_train DataFrames, a reusable folds object, notes/competition_overview.md, and an issues checklist for leakage, imbalance, and high-cardinality features.

Quick Start

Run tabpfn-explore on your train and test CSVs to generate cleaned DataFrames, a saved folds object, adversarial validation diagnostics, and an API cell budget check.

Frequently Asked Questions about tabpfn-explore

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I detect data leakage and distribution shift in Kaggle tabular datasets?▼

Adversarial validation detects data leakage and distribution shift by running train-vs-test classifiers to quantify differences and surface the specific features driving the separation between your training and testing data.

What is the best way to profile missing values and high-cardinality categoricals for classification competitions?▼

Exploratory data analysis profiles missingness, class balance, high-cardinality categoricals, duplicates, and high correlations to document dataset characteristics and guide your modeling choices for classification and regression tasks.

How do I set up reproducible cross-validation folds for tabular competition data?▼

Define and save reproducible cross-validation folds using StratifiedKFold, GroupKFold, or TimeSeriesSplit to generate consistent fold split indices that ensure robust model evaluation across your training data.

Can I check my TabPFN API cell budget constraints before starting model training?▼

You can verify TabPFN API cell budget constraints during the cross-validation scheme setup to prevent exceeding API limits before initiating any downstream modeling or API calls.

What outputs do I get from systematic exploratory data analysis on competition datasets?▼

Outputs include cleaned X_train/X_test/y_train DataFrames, a reusable folds object, adversarial validation AUC diagnostics, a competition overview notes file, and an issues checklist for leakage and imbalance.

Related Skills

Explore More Data & Analytics Skills→
dotlab-hqdotlab-hq

cudf-analytics

Perform GPU-accelerated groupby aggregations and statistical summaries on large CSV datasets.

Official
Intermediate
OlixIgnaciousOlixIgnacious
1

kaggle-validation

Detect distribution shifts and select fold types for Kaggle validation.

Community
Intermediate
dianapriordianaprior
5

tabpfn-regress

Generate TabPFN regression baselines with out-of-fold predictions and ensembled submissions.

Community
Advanced
Vishnugupta2711Vishnugupta2711
1

xgboost-lightgbm

Train XGBoost and LightGBM models on tabular datasets with early stopping and cross-validation.

Community
Advanced