|
import gradio as gr |
|
import boto3, json, os |
|
|
|
from dotenv import load_dotenv, find_dotenv |
|
_ = load_dotenv(find_dotenv()) |
|
|
|
aws_access_key_id = os.environ["AWS_ACCESS_KEY_ID"] |
|
aws_secret_access_key = os.environ["AWS_SECRET_ACCESS_KEY"] |
|
|
|
bedrock_runtime = boto3.client( |
|
aws_access_key_id = aws_access_key_id, |
|
aws_secret_access_key = aws_secret_access_key, |
|
service_name = "bedrock-runtime", |
|
region_name = "us-west-2" |
|
) |
|
|
|
|
|
def invoke(prompt): |
|
body = json.dumps({"prompt": "\n\nHuman: " + prompt + "\n\nAssistant: ", |
|
"max_tokens_to_sample": 300, |
|
"temperature": 1, |
|
"top_k": 250, |
|
"top_p": 0.999, |
|
"stop_sequences": ["\n\nHuman: "] |
|
}) |
|
modelId = "anthropic.claude-v2" |
|
accept = "application/json" |
|
contentType = "application/json" |
|
response = bedrock_runtime.invoke_model(body = body, modelId = modelId, accept = accept, contentType = contentType) |
|
response_body = json.loads(response.get("body").read()) |
|
return response_body["completion"] |
|
|
|
gr.close_all() |
|
demo = gr.Interface(fn=invoke, |
|
inputs = [gr.Textbox(label = "Prompt", lines = 1)], |
|
outputs = [gr.Textbox(label = "Completion", lines = 1)], |
|
title = "Generative AI - Language", |
|
description = "<a href='https://www.gradio.app/'>Gradio</a> UI using <a href='https://aws.amazon.com/bedrock/'>Amazon Bedrock</a> API with <a href='https://www.anthropic.com/'>Anthropic</a> Claude 2 foundation model") |
|
demo.launch() |
