Automate feature-engineering experiments on the statcast-gbm model.
npx skills add https://github.com/edpaget/bot-showalter --skill experiment-edpaget
Please help me install this Agent Skill. Skill: experiment Source: https://github.com/edpaget/bot-showalter/tree/main/.claude/skills/experiment Command: npx skills add https://github.com/edpaget/bot-showalter --skill experiment-edpaget
This Skill automates the process of feature-engineering experiments on the statcast-gbm model, improving its predictive accuracy and efficiency.
Run the experiment skill with the player type 'batter' and target 'slg', with a budget of 10 iterations: experiment batter slg --budget 10
High-intent search queries and answers about installing and using this skill.
Automate feature engineering for a statistical sports analytics model by running an experiment loop that identifies weaknesses, generates hypotheses, tests candidates, and logs results to validate winners.
Run feature engineering experiments on the statcast-gbm model by executing the experiment command with player type, target, and a set iteration budget to trigger the automated hypothesis and validation loop.
Automated hypothesis generation for predictive modeling works by diagnosing model weaknesses, using correlation screening to filter candidate features, and testing them with fast feedback tools to validate improvements.
Yes, you need access to the statcast-gbm model and its associated data to run experiment automation, as the Skill requires this baseline to identify weaknesses and validate new features.
Yes, you can limit iterations when testing feature engineering candidates by setting a specific budget parameter in the command, such as using a budget of 10 to constrain the experiment loop.
Automated correlation screening for machine learning feature selection filters candidate variables quickly before full testing, reducing wasted iterations and isolating the most promising predictive features for validation.
PuckAPIAutomate sports analytics pipelines for data collection, predictions, and drift detection.
bomberoxenviosdosruedasAnalyze sales pipeline health and revenue forecasts using historical data and velocity metrics.
CompSci-SquadPerform advanced analytics, predictive modeling, and statistical modeling for data science tasks.
zhangzhang-111-iDesign A/B tests, engineer features, and evaluate models with MLflow.
