We compare a compact fine-tuned LLM against a lightweight supervised model for parsing recipes from noisy web pages. The article outlines experiments with commercial LLMs, local open-source models, and cost-performance considerations.
This article discusses speeding up ML training loops by caching serialized features to reduce repeated data processing time. It covers loading cached features, memory profiling, and evaluating a Logistic Regression model with performance…
The article documents a recipe data extractor that underperformed, mislabeling buttons and duplicating ingredients. It outlines planned incremental improvements with feature enrichment to approach production readiness.
This post trains an initial supervised learning model on labeled recipe data, walks through the training pipeline and reports the first evaluation metrics. It explains feature extraction, pipeline construction, and highlights practical next steps to improve model accuracy and reduce common misclassifications.
A hands-on introduction to building a supervised learning pipeline using labeled recipe data from the author's old Recipe Folder project. This post explains the dataset, setup steps, and the plan for a multi-part series to train and improve models for extracting recipes from web pages.