SM_CHANNEL_TRAIN: A string representing the path to the directory containing data in the ‘train’ channel The following environment variables are set, following the format “SM_CHANNEL_”: Suppose you use two input channels, named ‘train’ and ‘test’, in the call to the Chainer estimator’s fit() method. These artifacts are compressedĪnd uploaded to S3 to the same S3 prefix as the model artifacts. Include checkpoints, graphs, and other files to save, not including model artifacts. SM_OUTPUT_DATA_DIR: A string representing the filesystem path to write output artifacts to. SM_NUM_GPUS: An integer representing the number of GPUs available to the host. These artifacts are uploaded to S3 for model hosting. SM_MODEL_DIR: A string representing the path to the directory to write model artifacts to. ![]() ![]() The training script is similar to a training script you might run outside of SageMaker, but youĬan access useful properties about the training environment through various environment variables, Your Chainer training script must be a Python 2.7 or 3.5 compatible source file.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |