English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Jupyter notebooks enable developers to interleave code snippets withrich-text and in-line visualizations. Data scientists use Jupyter notebook asthe de-facto standard for creating and sharing machine-learning basedsolutions, primarily written in Python. Recent studies have demonstrated,however, that a large portion of Jupyter notebooks available on publicplatforms are undocumented and lacks a narrative structure. This reduces thereadability of these notebooks. To address this shortcoming, this paperpresents HeaderGen, a novel tool-based approach that automatically annotatescode cells with categorical markdown headers based on a taxonomy of MLoperations, and classifies and displays function calls according to thistaxonomy. For this functionality to be realized, HeaderGen enhances an existingcall graph analysis in PyCG. To improve precision, HeaderGen extends PyCG'sanalysis with support for handling external library code and flow-sensitivity.The former is realized by facilitating the resolution of function return-types.The evaluation on 15 real-world Jupyter notebooks from Kaggle shows thatHeaderGen's underlying call graph analysis yields high accuracy (95.6%precision and 95.3% recall). This is because HeaderGen can resolve return-typesof external libraries where existing type inference tools such as pytype (byGoogle), pyright (by Microsoft), and Jedi fall short. The header generation hasa precision of 85.7% and a recall rate of 92.8%. In a user study, HeaderGenhelps participants finish comprehension and navigation tasks faster. To furtherevaluate the type inference capability of tools, we introduce TypeEvalPy, aframework for evaluating type inference tools with a micro-benchmark containing154 code snippets and 845 type annotations. Our comparative analysis on fourtools revealed that HeaderGen outperforms other tools in exact matches with theground truth.

Static Analysis Driven Enhancements for Comprehension in Machine Learning Notebooks