libraryofcelsus / aetherius_ai_assistant Goto Github PK

Traceback (most recent call last):
  File "C:\Users\Administrator\Documents\Dev\Aetherius_AI_Assistant\main.py", line 96, in <module>
    main_menu()
  File "C:\Users\Administrator\Documents\Dev\Aetherius_AI_Assistant\main.py", line 56, in main_menu
    function()
  File "C:\Users\Administrator\Documents\Dev\Aetherius_AI_Assistant\scripts/OpenAi_General_Chatbot.py", line 48, in OpenAi_General_Chatbot
    function()
  File "C:\Users\Administrator\Documents\Dev\Aetherius_AI_Assistant\scripts/OpenAi_General_Chatbot/GPT_4_Tasklist_Web_Scrape.py", line 496, in GPT_4_Tasklist_Web_Scrape
    int_conversation.append({'role': 'assistant', 'content': "MEMORIES: %s;\n%s;\n%s;\n\n%s'S INNER THOUGHTS: %s;\nHEURISTICS: %s;\nUSER MESSAGE: %s;\nWEBSCRAPE SAMPLE: %s\nIn a single paragraph, interpret the user, %s's message as %s in third person by creating an intuitive plan on what information needs to be researched from the webscrape, even if the user is uncertain about their own needs.;\nINTUITION: " % (db_search_4, db_search_5, db_search_12, bot_name, output_one, db_search_13, a, int_scrape, username, bot_name)})
NameError: name 'db_search_13' is not defined. Did you mean: 'db_search_1'?

Whenever I try to start a chat using the "GPT-4 Web Scrape" configuration I get this error.

First the web scraping is done and then it allows you to start the chat, that's when you enter something that you get the error.

Hi @libraryofcelsus,

We tried doing something similar (Vector db qa bot) and faced a ton of reliability issues from openai (Rate limiting, server timeouts, etc.).

Would recommend wrapping the openai calls with reliableGPT - it'll handle retries, model switches, etc. for you

from reliablegpt import reliableGPT
openai.ChatCompletion.create = reliableGPT(openai.ChatCompletion.create, ...)

Source: https://github.com/BerriAI/reliableGPT

Why during embedding generation, is gpt-3.5-turbo used instead of text-embedding-ada-002?

I am seeing things like this in my billing panel:

Local time: 8 jun 2023, 14:15
gpt-3.5-turbo-0301, 63 requests
18,792 prompt + 4,115 completion = 22,907 tokens

Local time: 8 jun 2023, 14:15
text-embedding-ada-002-v2, 4 requests
493 prompt + 0 completion = 493 tokens

That is definitely not normal, doing the embedding from GPT instead of Ada is a very high cost if you have to process up to 1M tokens as I usually have to do.

Reviewing the code, it has indeed been established that the embeddings are delegated to text-embedding-ada-002, but when it comes down to it, it is gpt-3.5 that is generating them.

First off, thank you for the comprehensive guide! Its a lot more detailed than I have seen for other projects on here and its always appriciated. Ive gotten into the habit of copy and pasting readmes into chatgpt and asking for a comprensive guide. I do have a few thoughts/suggestions/issues i am running into

• I am having trouble understanding what is meant by cadence in step 17. So I need an example of a speaking style but this is a typed thing correct? Do I just need a sample text? What if I'm not quite sure yet what I want their diction to be like? Is this 100% necessary? A little more info would be greatly appreciated maybe a sample would be nice.
• In training mode, would it possible to change the memories that are added directly?
• Is pinecone the only way to store the memories? I am not too into the whole deleting after 7 days thing, and while i'd be willing to pay like 10$/month for persistent memory but their cheapest plans seem to be 70$ which is obviously a bit outside my budget. I don't really quite understand exactly how pinecone works and if it provides essential functions to this but if possible, since this is being run on my local pc, itd be nice to be able to store all of this locally without having to worry about another service

Apologies if anything in here is dumb or bad form. I am a beginner to all this stuff. I dont expect a speedy reply or changes or anything like that. I'm always worried about annoying the nice developers that make their code freely available, so let me know if I have over stepped my bounds or anything

Users want access to as much AI as they can get, they dont want to manage 50 accounts, they want the fastest AI they want the cheapest AI, and you can provide all of that for them with this update.

in addition to or in place of integrating with any aggregators - Please integrate APIpie so devs can access them all from one place/subscription and plus it also provides:

-The most affordable, reliable and fastest AI available
-One API to access ~500 Models and growing
-Language, embedding, voice, image, vision and more
-Global AI load balancing, route queries based on price or latency
-Redundancy for major models providing the greatest up time possible
-Global reporting of AI availability, pricing and performance

Its the same API format as openai, just change the domain name and your API key and enjoy a plethora of models without changing any of your code other than how you handle the models list.

This is a win win for everyone, any new AI's from any providers will be automatically integrated into your stack with this one integration. Not to mention all the other advantages.