Skip to content

Commit

Permalink
Merge pull request #46 from UMLCloudComputing/ultralapse-patch-1
Browse files Browse the repository at this point in the history
docs: add information about Amazon Bedrock Knowledge Bases
  • Loading branch information
ultralapse authored Jun 14, 2024
2 parents 4bf297a + f3661d1 commit 1e5dede
Showing 1 changed file with 21 additions and 1 deletion.
22 changes: 21 additions & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -107,7 +107,7 @@ When the bot is not in use, the Lambda Function will not run, significantly savi

1. Install the tools listed in the Dependencies section of the README.md
2. Clone the repository.
3. Create an IAM user that can access AWS Lambda and Cloudformation. Create an access key that you'll use in the next step.
3. Create an IAM user that can access AWS Lambda and Cloudformation. Create an access key that you'll use in the next step. You can also put the same exact credentials for `BEDROCK_ID` and `BEDROCK_KEY` in the `.env` file
4. Run `aws configure` to setup your AWS credentials.
5. Go to discord.dev and create a new application.

Expand All @@ -128,6 +128,7 @@ Secret Key:
Public Key:
![image](https://github.com/UMLCloudComputing/rowdybot/assets/136134023/595f713f-c415-4b1d-937f-86929e0c5e00)


7. Save them in a `.env` file like the one below:

Example of `.env` file.
Expand All @@ -138,11 +139,30 @@ ID=<ID of bot>
LAMBDA_FUNC=<name of your lambda function (can be anything)>
BEDROCK_ID=<bedrock IAM user ID>
BEDROCK_KEY=<bedrock IAM user key>
KNOWLEDGE_BASE_ID=<Amazon Bedrock Knowledge Base ID. Put AZR9D11EGV if you don't have your own Knowledge Base>
```
8. Finally, run `cdk bootstrap` to setup the cdk project.

</details>

## 🪨 Setting up Amazon Bedrock

1. Head to your main AWS Dashboard and search for Amazon Bedrock. Click on Amazon Bedrock

![image](https://github.com/UMLCloudComputing/rowdybot/assets/136134023/26fdea83-2d4e-4a06-a4d1-e15071ec6b8e)

Click on Get Started

![image](https://github.com/UMLCloudComputing/rowdybot/assets/136134023/db7aa135-d3ea-494c-8048-c6e75f7c64ae)

Click on the Titan Models category and request access to Titan Text G1 - Premiere, Titan Text G1 - Lite, and Titan Text Embeddings v2
**If you're using the Cloud Computing Club account, then the necessary models have already been requested.**

![image](https://github.com/UMLCloudComputing/rowdybot/assets/136134023/a6b0b9c3-f5f2-402d-a41e-418e54f9aafb)

Now click on AWS Bedrock Knowledge Bases and create a Knowledge Base. You can use the Knowledge Base ID provided in "Setup" if you don't want to create one.
![image](https://github.com/UMLCloudComputing/rowdybot/assets/136134023/036664e0-ad62-43da-b5df-b9c19b92de36)

## 👉 Commands

<details>
Expand Down

0 comments on commit 1e5dede

Please sign in to comment.