Amazon Nova 모델에 대한 미세 조정 작업 생성 및 관리

콘솔에서 RFT 작업을 제출하려면 다음 단계를 수행합니다.

를 로 customizationType 설정하여 CreateModelCustomizationJob 요청을 보냅니다REINFORCEMENT_FINE_TUNING.

필수 필드: roleArn, baseModelIdentifier, customModelName, jobName, trainingDataConfig, outputDataConfig, rftConfig

요청 예제:

{
    "roleArn": "arn:aws:iam::123456789012:role/BedrockRFTRole",
    "baseModelIdentifier": "amazon.nova-2.0",
    "customModelName": "my-rft-model",
    "jobName": "my-rft-job",
    "customizationType": "REINFORCEMENT_FINE_TUNING",
    "trainingDataConfig": {
        "s3Uri": "s3://my-bucket/training-data.jsonl"
    },
    "customizationConfig": {
        "rftConfig" : {
            "graderConfig": {
                "lambdaGrader": {
                    "lambdaArn": "arn:aws:lambda:us-east-1:123456789012:function:function-name"
                }
            },
            "hyperParameters": {
                "batchSize": 64,
                "epochCount": 2,
                "evalInterval": 10,
                "inferenceMaxTokens": 8192,
                "learningRate": 0.00001,
                "maxPromptLength": 4096,
                "reasoningEffort": "high",
                "trainingSamplePerPrompt": 4
            }
        }
    },
    "outputDataConfig": {
        "s3Uri": "s3://my-bucket/rft-output/"
    }
}

Python API 샘플 요청:

import boto3

bedrock = boto3.client(service_name='bedrock')
    
# Set parameters
customizationType = "REINFORCEMENT_FINE_TUNING"
baseModelIdentifier = "arn:aws:bedrock:us-east-1::foundation-model/amazon.nova-2-lite-v1:0:256k"
roleArn = "${your-customization-role-arn}"
jobName = "MyFineTuningJob"
customModelName = "MyCustomModel"

customizationConfig = {
    'rftConfig' : {
        'graderConfig': {
            'lambdaGrader': {
                'lambdaArn': 'arn:aws:lambda:us-east-1:123456789012:function:function-name'
            }
        },
        'hyperParameters': {
            'batchSize': 64,
            'epochCount': 2,
            'evalInterval': 10,
            'inferenceMaxTokens': 8192,
            'learningRate':0.00001,
            'maxPromptLength': 4096,
            'reasoningEffort': 'high',
            'trainingSamplePerPrompt':4
        }
    }
}

trainingDataConfig = {"s3Uri": "s3://${training-bucket}/myInputData/train.jsonl"}
outputDataConfig = {"s3Uri": "s3://${output-bucket}/myOutputData"}

# Create job
response_ft = bedrock.create_model_customization_job(
    jobName=jobName, 
    customModelName=customModelName,
    roleArn=roleArn,
    baseModelIdentifier=baseModelIdentifier,
    customizationConfig=customizationConfig,
    trainingDataConfig=trainingDataConfig,
    outputDataConfig=outputDataConfig,
    customizationType=customizationType
)

jobArn = response_ft['jobArn']