extract-hyperparameters

Extract hyperparameters from research papers into YAML/JSON configuration files.

18|5|Updated Nov 3, 2025

Search Tags:#data science #machine learning #research papers #data preprocessing #training configuration #hyperparameters #model architecture

Ready to run

npx skills add https://github.com/mvillmow/ProjectOdyssey --skill extract-hyperparameters

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: extract-hyperparameters
Source: https://github.com/mvillmow/ProjectOdyssey/tree/main/.claude/skills/tier-2/extract-hyperparameters
Command: npx skills add https://github.com/mvillmow/ProjectOdyssey --skill extract-hyperparameters

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pdftotext, grep.

What problem does it solve?

Manually locating and documenting all hyperparameters from research papers for model training is a meticulous and time-consuming task, prone to errors.

Core Features & Use Cases

Parameter Extraction: Locates and documents all hyperparameters mentioned in research papers, including learning rates, batch sizes, and model configurations.
Configuration File Generation: Helps translate extracted parameters into an implementation-ready configuration file format (e.g., YAML, JSON).
Use Case: When reproducing results from a machine learning research paper, use this skill to quickly extract all relevant hyperparameters, such as learning rate, batch size, and optimizer type, to set up your training configuration.

Quick Start

Use the extract-hyperparameters skill to extract common ML hyperparameters like learning rate and batch size from 'paper.pdf'.

Frequently Asked Questions about extract-hyperparameters

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

FAQPage Schema

How do I extract hyperparameters from research papers?▼

Extract hyperparameters by parsing PDFs or text files to locate learning rates, batch sizes, epochs, momentum, weight decay, dropout, layer sizes, activation functions, normalization methods, optimizers, and preprocessing steps. The Skill identifies and documents these values in structured format ready for YAML or JSON configuration files.

Can I use this to reproduce machine learning results from published papers?▼

Yes. Extract all hyperparameters mentioned in research papers to set up training configurations that faithfully replicate the original work. This enables accurate reproduction of paper results by capturing architecture details, optimizer settings, and training parameters.

What hyperparameter values does this extract?▼

The Skill extracts learning rate, batch size, epochs, momentum, weight decay, dropout rates, layer sizes, activation functions, normalization techniques, optimizer type, and data preprocessing steps. It parses both structured and prose descriptions in PDFs and text files.

What file formats does hyperparameter extraction support?▼

Extract hyperparameters from PDF research papers and plain text files. The Skill uses pdftotext and grep to parse documents and locate parameter values and configuration details throughout paper content.

How do I compare hyperparameters across different research papers?▼

Extract hyperparameters from multiple papers into structured configuration format, then compare learning rates, batch sizes, model architectures, and optimizer settings side by side. This enables analysis of hyperparameter choices across works and informs tuning experiments.