AWS bedrock documentation change
Summary
Updated prerequisites for model customization jobs, adding Distillation as a new option and restructuring the dataset preparation instructions based on the chosen customization method.
Security assessment
The changes are focused on adding a new customization method (Distillation) and restructuring documentation for different customization types. There is no mention of security vulnerabilities, incidents, or specific security enhancements.
Diff
diff --git a/bedrock/latest/userguide/model-customization-prereq.md index 789e34e09..670da32e9 100644 --- a/bedrock/latest/userguide/model-customization-prereq.md +++ b/bedrock/latest/userguide/model-customization-prereq.md @@ -9 +9 @@ Before you can start a model customization job, you need to fulfill the followin - 1. Determine whether you plan to carry out a Fine-tuning or Continued Pre-training job and which model you plan to use. The choice you make determines the format of the datasets that you feed into the customization job. + 1. Determine the type of customization job (Distillation, Fine-tuning, or Continued Pre-training) and the model you plan to use. The choice you make determines the format of the datasets that you feed into the customization job. @@ -11 +11 @@ Before you can start a model customization job, you need to fulfill the followin - 2. Prepare the training dataset file. If the customization method and model that you choose supports a validation dataset, you can also prepare a validation dataset file. Follow the steps below in [Prepare the datasets](./model-customization-prepare.html) and then [upload](https://docs.aws.amazon.com/AmazonS3/latest/userguide/upload-objects.html) the files to an Amazon S3 bucket. + 2. Prepare the training dataset file. If the customization method and model that you choose supports a validation dataset, you can also prepare a validation dataset file. @@ -13 +13,5 @@ Before you can start a model customization job, you need to fulfill the followin - 3. (Optional) Create a custom AWS Identity and Access Management (IAM) [service role](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_terms-and-concepts.html#iam-term-service-role) with the proper permissions by following the instructions at [Create a service role for model customization](./model-customization-iam-role.html) to set up the role. You can skip this prerequisite if you plan to use the AWS Management Console to automatically create a service role for you. + * If you are planning to use Distillation, follow the steps in [Prerequisites for Amazon Bedrock Model Distillation](./prequisites-model-distillation.html) and then [upload](https://docs.aws.amazon.com/AmazonS3/latest/userguide/upload-objects.html) the files to an Amazon S3 bucket. + + * If you are planning to use Fine-tuning or a Continued Pre-training, follow the steps in [Prepare the datasets](./model-customization-prepare.html) and then [upload](https://docs.aws.amazon.com/AmazonS3/latest/userguide/upload-objects.html) the files to an Amazon S3 bucket. + + 3. Create a custom AWS Identity and Access Management (IAM) [service role](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_terms-and-concepts.html#iam-term-service-role) with the proper permissions by following the instructions at [Create a service role for model customization](./model-customization-iam-role.html) to set up the role. You can skip this prerequisite if you plan to use the AWS Management Console to automatically create a service role for you.