Finance-Domain-Expert-LLM-Model: Generative AI with AWS

Proof of concept (POC) for a domain expert model:

This model will be trained on a dataset of domain-specific knowledge in finance.

Training and Fine-Tuning a Large Language Model on AWS:

Meta Llama 2 7B foundation model Ver 1.0.0

Chat Optimized, Text Generation, LLAMA 2

Llama 2 is an auto-regressive language model that uses an optimized transformer architecture.

JumpStartModel

(model_id, model_version,) = ("meta-textgeneration-llama-2-7b","2.*",)

JumpStartEstimator

(model_id=model_id,  environment={"accept_eula": "true"},instance_type = "ml.g5.2xlarge")

Instance type for training job usage (EC2 instance with GPU for fine-tuning from AWS Service Quotas)

ml.g5.2xlarge 

AWS SageMaker IAM Role for ComputeFullAccess

AWS SageMaker Notebook

instance(ml.t3.medium)

platform(Amazon Linux 2, Jupyter Lab 3)

kernel(Python 3.10 Pytorch or Tensorflow)

Using the AWS platform and tools like,

AWS SageMaker (Python SDK and boto3, JumpStartModel, JumpStartEstimator)

AWS Rekognition

AWS EC2 

AWS S3 (s3://genaiwithawsproject2024/training-datasets/finance)

Amazon Bedrock

AWS PartyRock

AWS CodeWhisperer

AWS Cloud9

AWS Service Quotas