AWS SageMaker - training locally but deploying to AWS?

Question

I have a the following challenge with SageMaker:

I've downloaded one of the tutorial notebooks (https://github.com/awslabs/amazon-sagemaker-examples/blob/master/sagemaker-python-sdk/tensorflow_abalone_age_predictor_using_keras/tensorflow_abalone_age_predictor_using_keras.ipynb)

I ran the training locally (successfully) with the modifying the following line:

abalone_estimator = TensorFlow(entry_point='abalone.py',
                       role=role,
                       training_steps= 100,
                       evaluation_steps= 100,                                 
                       hyperparameters={'learning_rate': 0.001},
                       train_instance_count=1,
                       **train_instance_type='local'**)

abalone_estimator.fit(inputs)

I then wanted to deploy my model to AWS with the following line but it seems the SDK deploys it locally (it doesn't fail, I just see it running on my machine)

abalone_predictor = abalone_estimator.deploy(initial_instance_count=1, instance_type='ml.m4.xlarge')

Any tips on how to either fix it so it gets deployed to AWS or alternatively re-load my training model and deploy it to AWS from scratch?

Many thanks, Stefan

score 2 · Answer 1 · answered Apr 26 '18 at 18:00

2

Its easier to run the training again on SageMaker. Otherwise, here are the steps that you would have to do.

Take the checkpoint file generated during the training and convert them into tensorflow serving models.
Zip them in a specific format and upload to S3
Then create estimator as you have done above and do the inference.

If you want details on each of the specific steps above do let me know, but if your dataset is not too big, I would say just retrain on SageMaker.

answered Apr 26 '18 at 18:00

Raman

643
5
6

This is great, thanks Raman! Yes, details on how to do these 3 steps would be great (links are fine, just couldn't find much examples online). The reason is that while I'm trying with the standard demo model, my actual model & data set will be huge but not time sensitive so I'd like to train locally (e.g. overnight or longer) and then deploy the final model to SageMaker. – Stefan Apr 27 '18 at 09:47
How do I do it with other ML models like random forest/linear regressor, etc? – Eswar Aug 08 '19 at 10:18

AWS SageMaker - training locally but deploying to AWS?

1 Answers1

Linked