JosePizarro3 / NERxiv

91%

LAST BUILD ON BRANCH main

branch: main

CHANGE BRANCH

x

Reset

Sync Branches

• main
• 15-improve-prompting-with-oop-idea
• 17-fix-testing
• 20-add-extraction-from-prompt-to-structured-pydantic-model
• 22-define-full-system-and-method-model
• 23-add-class-diagrams
• 29-add-parameters-control-of-llm
• 30-store-chunks-in-hdf5
• 34-write-documentation
• 45-define-prompts-in-prompt_registry-for-all-the-sections-mentioned-add-only-the-retriever_query-and-some-empty-schema-basemodel-sections-add-a-normalize-method-and-a-base-class-common-to-all-sections
• 47-split-structuredprompts-for-sections-from-filters-to-filter-out-papers
• 49-change-run_xxxx-in-hdf5-for-the-key-in-the-prompt_registry-useful-to-later-implement-reusing-chunks-and-so-on-add-run_xxxx-under-the-group-of-the-prompt_registry-eg-raw_llm_answers-dft-run_0001-run_0002-etc
• 50-add-chunks-reading-logic
• 52-work-on-the-dft-section-details-with-the-help-of-a-bunch-of-papers-and-use-chatgpt-if-necessary
• 54-with-the-new-dft-schema-identify-examples-and-point-to-main-fields-in-the-schema-split-common-fields-in-other-base-models-eg-software-for-the-program-name-and-version
• 59-add-multiple-prompting-scripts-and-averaging-response-logic
• 67-polished-documentation
• add_font_woff
• bump_pyrxiv_dep
• copilot/add-chunk-hash-functionality
• copilot/add-rag-extractor-agent-docs
• copilot/fix-agent-testing-coverage
• copilot/fix-mkdocs-live-reload-issue
• copilot/fix-mkdocs-page-deployment
• copilot/fix-pipeline-issues-and-docs
• copilot/minimize-hallucinations-llm
• copilot/sub-pr-55
• copilot/sub-pr-62
• feature/rag-extractor-agent
• fix_mkdocs_serve
• fix_tutorial_rag_extractor_agent
• hotfix_docs
• improved_dft_retrieval
• minor_fixes
• revert-61-copilot/rebase-and-improve-agents
• utils_to_move_files
• v0.0.0
• v0.0.1
• v0.1.0
• v1.0.0
• v1.0.1

Build # 19300660794

Build Type

push

github

Committed by web-flow

Commit Message

Fix tutorial rag extractor agent (#73)

Fix things that do not work after testing

Run Details

495 of 541 relevant lines covered (91.5%)

0.91 hits per line