wangtao
/
autogen
Not watched
1
0
Fork 0
Code
Releases 127 Wiki evaluate Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain HPC
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
1655 Commits
199 Branches
282 MB
21 Download
 
 
 
 
 
 
Tree: 63bfe837ec
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '63bfe837ec'
${ noResults }
autogen/samples/tools/profiler
Gagan Bansal 63bfe837ec Add caching 2 years ago
..
profiler Add caching 2 years ago
README.md Update readme 2 years ago
demo.py Add initial profiler 2 years ago
requirements.txt Improve readme and printing 2 years ago
setup.py Improve and add CLI 2 years ago

README.md

aprofile

This package introduces a new functionality to profile chat messages. It includes a Profiler class that identifies the state of a chat message based on predefined states.
It also introduces a command called aprofile which can be used to profile chat messages loaded from files, json str input, and AutoGen Bench style console logs.

Installation

To install the package, clone the repository and install the dependencies.

# clone the correct repo/branch
# git clone git@github.com:microsoft/autogen.git
# git checkout ct_webarena
cd samples/tools/profiler
pip install -e .

Demo: API

A demonstration of how to use the Profiler class to profile a list of chat messages is provided in demo.py.

Demo: Command-Line Interface

The aprofile CLI can accept various input formats.

# for printing help
aprofile --help
# input: json string
aprofile --json [{"source": "user", "content": "plot a chart"}]
# input: path to a json
aprofile --file chat_history.json
# input: AGBench console log
aprofile --agbconsole <path to AGBench output>/console_log.txt

The command utility will print sources observed in the chat and their high-level states.

Delete task

Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.

cancel
confirm

This is a mirror of AutoGen from GitHub. AutoGen is a framework that enables the development of LLM applications using multiple agents that can converse with each other to solve tasks.

Python SVG Jupyter Notebook C# TSX other