Package: mergen 0.2.1

Altuna Akalin

mergen:AI-Driven Code Generation, Explanation and Execution for Data Analysis

Employing artificial intelligence to convert data analysis questions into executable code, explanations, and algorithms. The self-correction feature ensures the generated code is optimized for performance and accuracy. 'mergen' features a user-friendly chat interface, enabling users to interact with the AI agent and extract valuable insights from their data effortlessly.

Authors:Altuna Akalin [aut, cre], Jacqueline Anne Jansen [aut]

mergen_0.2.1.tar.gz
mergen_0.2.1.tar.gz(r-4.5-noble)mergen_0.2.1.tar.gz(r-4.4-noble)
mergen_0.2.1.tgz(r-4.4-emscripten)mergen_0.2.1.tgz(r-4.3-emscripten)
mergen.pdf |mergen.html✨
mergen/json (API)
NEWS

# Installmergen in R:

install.packages('mergen',repos = c('https://cran.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/bimsbbioinfo/mergen/issues

13 exports 0.23 score 35 dependencies 207 downloads

Last updated 8 days agofrom:516891211c

Exports:check_install clean_code_blocks executeCode extractCode extractFilenames extractInstallPkg fileHeaderPrompt promptContext runCodeInResponse sampleResponse selfcorrect sendPrompt setupAgent

Dependencies:askpass assertthat base64enc BiocManager bslib cachem cli curl digest evaluate fastmap fontawesome fs glue highr htmltools httr jquerylib jsonlite knitr lifecycle magrittr memoise mime openai openssl R6 rappdirs rlang rmarkdown sass sys tinytex xfun yaml

Introduction to mergen

Rendered frommergen.Rmdusingknitr::rmarkdownon Jun 28 2024.

Last update: 2023-12-09
Started: 2023-12-09

Citation

Readme and manuals

Help Manual

Help page	Topics
Check and Install R Package	check_install
Clean code blocks returned by the agent	clean_code_blocks
execute code	executeCode
extract the code and text from the text returned by LLM agent	extractCode
Extract file names from user prompt	extractFilenames
extract package names and install them	extractInstallPkg
Extract file headers from files in prompt	fileHeaderPrompt
Predefined prompt contexts for prompt engineering	promptContext
Executes the code in the received response from the agent	runCodeInResponse
Sample more solutions when non-executable code returned by the agent	sampleResponse
Self correct the code returned by the agent	selfcorrect
Send a prompt to a specified language model agent and return the response.	sendPrompt
set up an online LLM API for subsequent tasks	setupAgent