![]() We verified its effectiveness on a large-scale codebase with ~41k repositories.Įxperimental results showed that CodeMatcher achieves an MRR of 0.60, The candidate code snippet sequentially matched the important words in a query. Query words on the codebase that is indexed by the Elasticsearch tool, andįinally reranks a set of returned candidate code according to how the tokens in Irrelevant/noisy ones, then iteratively performs fuzzy search with important CodeMatcher first collects metadata for query words to identify Indexing technique in the IR-based model to accelerate the search response time ![]() In this paper, we proposed an IR-based modelĬodeMatcher that inherits the advantages of DeepCS, while it can leverage the Understanding irrelevant/noisy keywords and capturing sequential relationshipsīetween words in query and code. Two major advantages of DeepCS are the capability of The relationship between pairs of code methods and corresponding natural AnĮarly successful deep learning based model DeepCS solved this issue by learning Search, but they fail to connect the semantic gap between query and code. ![]() Years, researchers proposed many information retrieval based models for code Download a PDF of the paper titled CodeMatcher: Searching Code Based on Sequential Semantics of Important Query Words, by Chao Liu and 5 other authors Download PDF Abstract: To accelerate software development, developers frequently search and reuseĮxisting code snippets from a large-scale codebase, e.g., GitHub. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |