A package to evaluate the models.
Project description
pycvtools
Functionality of the pyevaltool
- Automated testing of the bot response deployed gpt
pip install pyevaltool
To run
from pyevaltool import *
For GPT test -
url = '' # Add the frontend url
token = '' # add your token or key - Apim-Subscription-Key or Bearer token
# change the chat completion format if required.
# This code will work only if the question asked is at chat_completion_format['messages'][0]["content"]
chat_completion_format = {'chatId': '', 'siteInfo': {'domain': ''}, 'messages': [{'role': 'user', 'content': ''}], 'model': '', 'stream': False}
answer_loc = "['choices'][0]['messages'][0]['content']" #in string format
input_file = " " # Here you add your input file link
gpt_test(url,token,chat_completion_format,answer_loc,input_file,output_file = "output.csv" )
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
pyevaltool-0.1.5.tar.gz
(3.6 kB
view hashes)
Built Distribution
Close
Hashes for pyevaltool-0.1.5-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | c4463e204a3bbc6c393bda3695be7e2370e81abed20bc7b98e182a99185193ff |
|
MD5 | aa84b3e9dbedf39567fe2dceb1123110 |
|
BLAKE2b-256 | 2bd2a91a00f265b745754c7c47ccf3204d744bf784fba1719ec4908e5a407c91 |