How to load model programmatically? #1094
Description
I notice that whenever I run a new batch transform jobs, it will create a new model and saves it.
I can see all the models from my batch transform jobs in my AWS Sagemaker Dashboard/inference/models
Here is the script that I run
sagemaker_model = MXNetModel(model_data = 's3://' + sagemaker_session.default_bucket() + '/model/yolo_object_person_detector.tar.gz',
role = role,
entry_point = 'entry_point.py',
py_version='py3',
framework_version='1.4.1',
sagemaker_session = sagemaker_session)
transformer = sagemaker_model.transformer(instance_count=1, instance_type='ml.m4.xlarge', output_path=batch_output)
transformer.transform(data=batch_input, content_type='application/x-image')
transformer.wait()I've looked into the source code for declaration of class MXNetModel
class MXNetModel(FrameworkModel):
"""An MXNet SageMaker ``Model`` that can be deployed to a SageMaker ``Endpoint``."""
__framework_name__ = "mxnet"
_LOWEST_MMS_VERSION = "1.4"
def __init__(
self,
model_data,
role,
entry_point,
image=None,
py_version="py2",
framework_version=MXNET_VERSION,
predictor_cls=MXNetPredictor,
model_server_workers=None,
**kwargs
):
...But I am not seeing anywhere where I can simply load the MXNetModel object using a URL Endpoint to the models in my dashboard.
If I go to console and click one of those models, I can see a button for Create batch transform job, so I know internally this is possible. But I can't find anything on the docs to do it programmatically.
Also as a side question:
How many models does the Free tier provide? In the free tier page: https://aws.amazon.com/sagemaker/pricing/ it just says the number of hours, but not necessarily the number of models