skills.rest
thbraet
thbraetCOMMUNITY
Explore all skills by this author →
GitHub

generate-submission

Generate a Kaggle submission CSV from the best model and formatted test data.

Updated Mar 20, 2026
Search Tags:#validation#pandas#submission#mlflow#kaggle#model-predictions
One-click install
npx skills add https://github.com/thbraet/claude-template --skill generate-submission
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: generate-submission
Source: https://github.com/thbraet/claude-template/tree/main/skills/generate-submission
Command: npx skills add https://github.com/thbraet/claude-template --skill generate-submission

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill automates the creation of a competition-ready Kaggle submission file from the best-performing model by loading the model, running predictions on the test data, and formatting the results into a CSV that conforms to the submission template.

Core Features & Use Cases

  • Load a model from a specified path, an MLflow run, or automatically pick the best model according to evaluation metrics.
  • Predict on the final, formatted test set and generate a properly structured submission CSV.
  • Validate the submission against the reference template (column names, order, and allowed value ranges) and save it alongside a generated notebook for traceability.

Quick Start

Run the skill with an optional model path, MLflow run ID, or 'best' to generate a competition-ready submission file.

Frequently Asked Questions about generate-submission

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

FAQPage Schema
How do I generate a Kaggle submission file from my best machine learning model?▼

To generate a Kaggle submission file, you load your best model, run predictions on the formatted test data, and save the results as a CSV conforming to the competition's submission template.

How does MLflow integration work when creating a submission CSV?▼

MLflow integration allows you to specify an MLflow run ID to load the corresponding model, predict on the test dataset, and automatically format the output into a validated competition submission file.

What format does my test data need to be in for Kaggle submission generation?▼

Your test data must be formatted and located at data/processed/test_formatted.csv. You also need a submission template to validate column names, order, and allowed value ranges for the output CSV.

Can I automatically select the best model to run predictions for my Kaggle competition?▼

Yes, you can pass 'best' to the skill to automatically select the top-performing model based on evaluation metrics, then predict on the test set and generate the final submission CSV.

How are Kaggle submissions validated against the competition template?▼

Submissions are validated by checking the generated CSV against a reference template to ensure correct column names, column order, and allowed value ranges before saving the final file.

What's the best way to ensure traceability when generating model predictions for Kaggle?▼

To ensure traceability when generating model predictions, the skill saves the final submission CSV alongside a generated notebook, linking the model loading and prediction steps to the output.

Related Skills

Explore More Data & Analytics Skills→
thbraetthbraet

data-lineage

Trace data lineage from raw data to model inputs and output a lineage.md report.

Community
Advanced
OlixIgnaciousOlixIgnacious
1

kaggle-grandmaster

Plan Kaggle competition strategies for feature engineering and model optimization.

Community
Advanced
dkyazzentwatwadkyazzentwatwa
86

model-comparison-tool

Compare machine learning models using cross-validation and performance metrics.

Community
Intermediate
zangxin75zangxin75

dataset-finder

Search, download, and manage datasets from Kaggle, Hugging Face, UCI ML, and Data.gov.

Community
Intermediate