Problem

Recent versions of openai lib require the OPENAI_API_KEY on openai.OpenAI() init.

This cascades down to bambooai on the following line (see traceback below):

Python 3.11.0 (main, Oct 24 2022, 18:26:48) [MSC v.1933 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import bambooai
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Users\USER\AppData\Roaming\Python\Python311\site-packages\bambooai\__init__.py", line 2, in <module>
    from .bambooai import BambooAI
  File "C:\Users\USER\AppData\Roaming\Python\Python311\site-packages\bambooai\bambooai.py", line 15, in <module>
    from . import models, prompts, func_calls, qa_retrieval, google_search, reg_ex, log_manager, output_manager, utils
  File "C:\Users\USER\AppData\Roaming\Python\Python311\site-packages\bambooai\google_search.py", line 10, in <module>
    openai_client = openai.OpenAI()
                    ^^^^^^^^^^^^^^^
  File "C:\Users\USER\AppData\Roaming\Python\Python311\site-packages\openai\_client.py", line 104, in __init__
    raise OpenAIError(
openai.OpenAIError: The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable

Suggested solution

Since google_search.py may not be needed, maybe we can do a conditional import.

I will try to implement a fix on the weekend following best practices (conditional imports usually are not).

Problem

When code corrections are triggered, the user is left waiting without any feedback on CLI about current status of the process (image below).

Solution

Output from the LLMs while correcting the code in between corrections would prevent the user from thinking the process has halted or crashed (green region in image below).

This is particularly troublesome doing inference on slow setups (such as local LLMs on laptops, like Llama3 8b).

How should we handle excel files with multiple sheets. I have tried and I am getting error- list object has no attribute 'columns'

Hi, there first of all this is an amazing project which you have built

But it seems that for each query even the basic ones token usage is in 40-50k
fopr queries based on personal .csv data (.csv data record contains 30 records only0
Any suggestions on how to reduce the token usage

Hi @pgalko,

Whenever I ask another question it seems to not take in the context of the previous question and answer and gives me a fresh new response.

I'd like to have memory enabled so that I can have a conversation with it.

Note - I am running the Bamboo AI Agent in a for-loop to enable memory
And also have set max_conversations=2.

This is my configuration
bamboo = BambooAI(df,llm='gpt-3.5-turbo-0613',max_conversations=2, debug=False, vector_db=False, exploratory=False, llm_switch=False, search_tool=False)

for _ in range(0,2):
user_input = input("Dear User Please Enter your query: ")
#How many rows are present from the dataset
bamboo.pd_agent_converse(user_input)

I am sure you have some solution, am I missing something?
Thank you.

How should I configure/use the model from Azure Open AI? to use this application

Hi @pgalko
Id really love using the Code-Llama LLM, I have just got access to it.
I need to know a few things

1. How can I set up with Code-Llama any example would be wonderful
2. What is the Minimum CPU requirement for a smooth running of the application with Code-Llama
3. Can we use Google Colab and add the Code-Llama model to it and run?