Metadata-Version: 2.1
Name: agenttobenamed
Version: 0.1.13
Summary: 
Author: poludmik
Author-email: michael.poludin@gmail.com
Requires-Python: >=3.9,<3.13
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Requires-Dist: langchain (>=0.0.336,<0.0.337)
Requires-Dist: matplotlib (>=3.8.1,<4.0.0)
Requires-Dist: numpy (>=1.26.2,<2.0.0)
Requires-Dist: openai (>=1.3.2,<2.0.0)
Requires-Dist: openpyxl (>=3.1.2,<4.0.0)
Requires-Dist: pandas (>=2.1.3,<3.0.0)
Requires-Dist: scikit-learn (>=1.3.2,<2.0.0)
Requires-Dist: scipy (>=1.11.3,<2.0.0)
Description-Content-Type: text/markdown

# A minimalistic LLM agent for Exploratory Data Analysis (EDA) using pandas library
## *Task*: given a CSV or a XLSX file, respond to user's query about this table by generating a python code and executing it.
* Using LangChain routers/tagging, a prompt type will be established: to produce a plot or to compute the numerical values.
* The GPT model breaks down the user's request into several subtasks.
* The LLM generates code based on these subtasks.
* The code is executed using the python REPL.
* If the REPL throws an error, GPT will begin debugging until resolved.
* The result is the string output of the code 'print(result)' + optionally, saved image with a plot.

**Query example:** 'Find correlation between GDP and happiness index, subtract 0.4 and multiply by 1e6.'

![alt text](https://github.com/poludmik/AgentToBeNamed/blob/master/README/AgentScreenshot.png?raw=true)

