thewickedkarma / amadeus

A Telegram Chatbot written in Python3 making use of OpenAI's GPT3 model (Da Vinci Engine). You can fork and deploy this repo on heroku.

Hi , I'm Ankit

A student from India, passionate about programming, Cyber Security, AI and Machine Learning.

Glad to see you here!

• 🌱 I’m currently learning C++ and various python libraries(OpenCV, TensorFlow, PyTorch). 
  

• ❤️ Open source lover and linux enthusiast.
  

• 🏹 I’m always looking forward to contribute to open source.
  

• 💬 Ask me about nothing, I am lazy.
  

• 📫 How to reach me: By teleportation...lol
  

• 😄 Pronouns: He/him
  

• ⚡ Fun fact: I don't know who you are 😶
  

• 💿 I use Arch btw 
  

Now-Playing/Recently Played 🎧

Some silly Stats 😗

    

Most Popular repo:

文 Languages and Tools 🛠

Languages:

Web Development:

Tools and Database:

Editors:

Operating Systems:

Trophies:

Contact:

amadeus's Issues

Error "You exceeded your current quota, please check your plan and billing details."

Not sure what's the error. The quote is not exceeded yet. Probably API request is outdated?

Not Getting multiline reply

When the API is replying with multiline text, the bot only showing first line.

Message text is empty

when asked complex question, bot responses with this reply: "Message text is empty"
also sometime it doesnt print the whole response. for example if response has multiple paragraph it would only print the first line.

P.S.; i tried text-davinci-003 model but same problems

"I'm sorry, I don't know how to help you with that."

I borrowed some of your code for my own TG bot (thanks for making this btw!!) to do some experiments with GPT-3. I'm wondering how I can adjust the parameters for temperature, frequency, etc. in the code itself

2 questions

1. i notice Amadeus often responds with "I'm sorry, I don't know how to help you with that" a lot more than inputting the same question into openai's playground. Do you know why this might be the case?

2. and this may solve my first question, but specifically I'm asking Amadeus a lot of data-related questions, so I'd like to adjust the temperature setting to be lower. Using the playground, I can grab this code but unsure where to include it in my local version of Amadeus:

response = openai.Completion.create(
  model="text-davinci-002",
  prompt="prompt input",
  temperature=0,
  max_tokens=100,
  top_p=1,
  frequency_penalty=0,
  presence_penalty=0
)

I think turning temperature way lower would help solve my issues, but curious if you've experienced this yourself

error filter

Traceback (most recent call last):
File "/Users/ethicdeveloper/Desktop/ChatGPT-Telegram-Bot/harshitethic.py", line 1, in
from telegram.ext import Updater, CommandHandler, MessageHandler, Filters
ImportError: cannot import name 'Filters' from 'telegram.ext' (/Users/ethicdeveloper/Library/Python/3.11/lib/python/site-packages/telegram/ext/init.py)