Amazon Neptune · Schema
CreateModelTrainingJobRequest
CreateModelTrainingJobRequest schema from Neptune
AWSDatabaseGraph DatabaseGremlinNeptuneProperty GraphRDFSPARQL
Properties
| Name | Type | Description |
|---|---|---|
| id | string | Unique identifier for the job. |
| dataProcessingJobId | string | Job ID of the completed data processing job. |
| trainModelS3Location | string | S3 location for model artifacts output. |
| previousModelTrainingJobId | string | Job ID of a previous training job for incremental training. |
| sagemakerIamRoleArn | string | |
| neptuneIamRoleArn | string | |
| modelName | string | The model type to train: rgcn (relational graph convolutional network), transe, distmult, rotate, or custom. |
| baseProcessingInstanceType | string | ML instance type for data preparation step. |
| trainingInstanceType | string | ML instance type for the training step. |
| trainingInstanceVolumeSizeInGB | integer | Disk volume size for training instance in GB. |
| trainingTimeOutInSeconds | integer | Training job timeout in seconds. |
| maxHPONumberOfTrainingJobs | integer | Maximum total training jobs for hyperparameter tuning. Minimum 10 recommended for meaningful results. |
| maxHPOParallelTrainingJobs | integer | Maximum parallel training jobs. |
| subnets | array | |
| securityGroupIds | array | |
| volumeEncryptionKMSKey | string | |
| s3OutputEncryptionKMSKey | string | |
| enableInterContainerTrafficEncryption | boolean | |
| enableManagedSpotTraining | boolean | Whether to use EC2 spot instances for training. |
| customModelTrainingParameters | object | Custom model training configuration. |