Thanks for the content. It would be helpful to link to previous videos and a high-level flowchart on how these need to be setup. (Sorry if you've already done this) For Example: step1: download Autogen packages step2: programming steps in IDE step3: getting/Adding API keys, and explaining how they will be referenced in individual agents (when we're mixing and matching free and paid models inside agents for max impact) step4: repurposing agents in a different workflow stepx: ensuring max optimization of prompts/minimal exhaustion of tokens
Hey thank you! Do you mind posting your code or maybe sending it in a link where it would be located? Might be easier to see it and help figure out the issue! Sorry you're having an issue, I will do my best to help
Random question: where does Microsoft get these Amazonian sounding names? They had a media player called Zune. and their cloud is called Azure...pronounced almost like ASHER but with a Z sound. wtf are they thinking with these names?
Traceback (most recent call last): File "c:\Users\User\Documents\app.py", line 1, in import autogen ModuleNotFoundError: No module named 'autogen'..... So I've donr pip install pyautogen and pymemgpt, ive uninstalled then, reinstalled them. I get the constant error on the first line saying there is no module named autogen... Could you PLEASE show us how to compile from source??? Its the only thing i can think of...
Hey sorry you're having so much trouble with this. Yes that is a good idea. So I have ran into this before, and honestly all I remember doing is just re-typing the import line. Sounds weird, but try that. Also I will create a separate project and clone my repo and see if I get the error and what I can do to fix. I will keep you posted on this!
Do I need to pay for the API subscription to make this work? because I pay the normal sub and it appears like I do have an API Key. (But i can't use gpt-4 with it) Is there a way that I can rig up like 3 or 4 different local llms and put gpt-3.5 turbo as the lead, give them each a role? For some reason I still can't get memgpt or autogen to work on my machine, and ive tried several times this last week for over an hour each. I've just now deleted and fresh installed both visual studio and anaconda so maybe itll work now......?
Hey, I think I need to make a fresh project since they have had updates and release a video on integrating them together again. This time I will have multiple llm's together to show exactly how it works. But yes, so if you use AutoGen then you will pay (I only have a normal sub as well, I don't pay for GPT-4). Even for 3.5-turbo as far as I know (I will verify) you have to pay, but its cheaper. So I put like $10 when I log in to open 'platform.openai.com' and login to my account and go to billing. There is a section on the left side for creating API Keys (you can create multiple). To answer your role question, yes this is possible! I have a good bit to cover with new stuff integrating into AutoGen. I am planning on opening up a discord for everybody as well so we can create a community to help each other and it may be easier to chat and fix issues.
Thank you, I appreciate that! I hope they are helping, I have more AutoGen integrations coming out. And unfortunately with the new tech things can change, API calls are likely to change (as MemGPT already has). So I will have 'newer' videos with updates to make it work. Nature of it. Yes I have an email in my about page
Thanks for the content.
It would be helpful to link to previous videos and a high-level flowchart on how these need to be setup.
(Sorry if you've already done this)
For Example:
step1: download Autogen packages
step2: programming steps in IDE
step3: getting/Adding API keys, and explaining how they will be referenced in individual agents (when we're mixing and matching free and paid models inside agents for max impact)
step4: repurposing agents in a different workflow
stepx: ensuring max optimization of prompts/minimal exhaustion of tokens
Just found this TH-cam channel and I really like the content! Great work Tyler!
Hey thank you I appreciate it! 👍
Couldn't have come at a better time!
Do you happen to know how to stop the azure openAI api and url not found warning?
Keep up the great content man.
Hey thank you! Do you mind posting your code or maybe sending it in a link where it would be located? Might be easier to see it and help figure out the issue! Sorry you're having an issue, I will do my best to help
Random question: where does Microsoft get these Amazonian sounding names? They had a media player called Zune. and their cloud is called Azure...pronounced almost like ASHER but with a Z sound. wtf are they thinking with these names?
Hey your guess is as good as mine, I guess their PR or HR has reasons lol. 🤷🤔
Traceback (most recent call last):
File "c:\Users\User\Documents\app.py", line 1, in
import autogen
ModuleNotFoundError: No module named 'autogen'.....
So I've donr pip install pyautogen and pymemgpt, ive uninstalled then, reinstalled them. I get the constant error on the first line saying there is no module named autogen... Could you PLEASE show us how to compile from source??? Its the only thing i can think of...
Hey sorry you're having so much trouble with this. Yes that is a good idea. So I have ran into this before, and honestly all I remember doing is just re-typing the import line. Sounds weird, but try that. Also I will create a separate project and clone my repo and see if I get the error and what I can do to fix. I will keep you posted on this!
Do I need to pay for the API subscription to make this work? because I pay the normal sub and it appears like I do have an API Key. (But i can't use gpt-4 with it) Is there a way that I can rig up like 3 or 4 different local llms and put gpt-3.5 turbo as the lead, give them each a role? For some reason I still can't get memgpt or autogen to work on my machine, and ive tried several times this last week for over an hour each. I've just now deleted and fresh installed both visual studio and anaconda so maybe itll work now......?
Hey, I think I need to make a fresh project since they have had updates and release a video on integrating them together again. This time I will have multiple llm's together to show exactly how it works. But yes, so if you use AutoGen then you will pay (I only have a normal sub as well, I don't pay for GPT-4). Even for 3.5-turbo as far as I know (I will verify) you have to pay, but its cheaper. So I put like $10 when I log in to open 'platform.openai.com' and login to my account and go to billing. There is a section on the left side for creating API Keys (you can create multiple). To answer your role question, yes this is possible! I have a good bit to cover with new stuff integrating into AutoGen.
I am planning on opening up a discord for everybody as well so we can create a community to help each other and it may be easier to chat and fix issues.
Tyler. Awesome content. Im CEO of a fintech, and super helpful vids feeding into team. Whats best way to contact you outside this channel?
Thank you, I appreciate that! I hope they are helping, I have more AutoGen integrations coming out. And unfortunately with the new tech things can change, API calls are likely to change (as MemGPT already has). So I will have 'newer' videos with updates to make it work. Nature of it. Yes I have an email in my about page