Skip to content

AWS Bedrock Model Registry Integration

Overview

IdeaWeaver integrates with AWS Bedrock for model hosting and deployment. This guide shows you how to use AWS Bedrock with IdeaWeaver for model management and inference.

Setup

Train and Deploy a Model

source ideaweaver-env/bin/activate
ideaweaver train \
  --model sshleifer/tiny-distilbert-base-cased \
  --dataset ./datasets/training_data.csv \
  --deploy-to-bedrock \
  --aws-region us-east-1 \
  --aws-access-key-id <your-access-key> \
  --aws-secret-access-key <your-secret-key>

Example Output

🤗 Using model: sshleifer/tiny-distilbert-base-cased
🚀 Starting model training...

============================================================
🎉 TRAINING SUMMARY
============================================================
📂 Model Path:           ./my-model
🤖 Base Model:           sshleifer/tiny-distilbert-base-cased
📊 Dataset:              ./autotrain_projects/my-model

📊 KEY PERFORMANCE METRICS
----------------------------------------
📉 Final Train Loss:     1.0986
🎯 Overall Accuracy:     20.0%

============================================================
✨ Training completed successfully! Model is ready for use.
============================================================

✅ Training completed successfully!
📁 Model saved to: ./my-model
🚀 Deploying model to AWS Bedrock...
✅ Model deployed successfully to AWS Bedrock

AWS Bedrock Console Example

Once your model is deployed, you can view it in the AWS Bedrock Console:

AWS Bedrock Console Screenshot

Features

  • Model hosting
  • Serverless deployment
  • Auto-scaling
  • Cost optimization
  • Security features
  • Monitoring tools

Configuration

Required Parameters

  • --deploy-to-bedrock: Flag to enable AWS Bedrock deployment
  • --aws-region: AWS region for deployment
  • --aws-access-key-id: AWS access key
  • --aws-secret-access-key: AWS secret key

AWS Credentials Setup

  1. Go to AWS IAM Console
  2. Create a new IAM user
  3. Attach necessary permissions
  4. Generate access keys
  5. Use keys in your training command

Best Practices

  1. Security
  2. Use IAM roles
  3. Enable encryption

  4. Monitor access

  5. Cost Management

  6. Set up budgets
  7. Monitor usage

  8. Optimize resources

  9. Deployment

  10. Use appropriate instance types
  11. Configure auto-scaling
  12. Set up monitoring

Troubleshooting

Common Issues

  1. Authentication Errors
  2. Verify AWS credentials
  3. Check IAM permissions

  4. Ensure proper access

  5. Deployment Issues

  6. Check region availability
  7. Verify resource limits
  8. Monitor CloudWatch logs

Debug Mode

Enable verbose output for debugging:

ideaweaver train \
  --model sshleifer/tiny-distilbert-base-cased \
  --dataset ./datasets/training_data.csv \
  --deploy-to-bedrock \
  --aws-region us-east-1 \
  --verbose

Resources