AI Training

Dall.e == visual creating tool, text to IImage model developed by OpenAI

 

GAN (Generative adversarial network)

 

 

Chat GPT

G = generative

P = Pretraining ( Being Fucntion and interact with models)

T = Transformation

 

 

 

we are alreadying working onto chatGPT3 and large language models to help increase our productivity

1. Creating Specific Databases

-> engineer can search for specific topics, and ChatGPT model is helping to make a decision and productivity

 

Conclusion

AI model is only good which is only for the data they already trained on

 

 

 

 

 

RAG (Retrieval-Augmented Generation) - allow us query our own documentation

LLM (Large Language Model)

 

create the Vector embedding function -> embedding data into the database

 

why do we do this?

vector embedding is the key part of the puzzle

we can map information into a space that retains semantic meaning of text , images and etc

 

semantic querying **

 

GPT-4 Model is the language model to transform input to answer

 

 

GPTs

Allow the user to customize the ChatGPT4, through text prompt and user data information

 

 

 

Copilot

AI tool which is integrated with office 365

 

Creating a document with prompt with provide proper document what I want to summarize

 

Secure, well define context, high-quality contents

We can ask copilot that search latest email/chat/files or any other in office platform. It can summarize & provide right information what I want to deal with

 

Generating a professional reply as well.

 

It helps to save time and being more productive

 

Copilot can access my shared onedrive & any other place I have an access (Sharepoint as well)