[tensowflow][mask_rcnn] mask-rcnn-experiment-trails.ipynb says it uses s3 but actually uses EFS

https://github.com/aws/amazon-sagemaker-examples/blob/master/advanced_functionality/distributed_tensorflow_mask_rcnn/mask-rcnn-experiment-trials.ipynb

The notebook in the introduction lists steps and mentions:

"Concretely, we will describe the steps for SagerMaker Experiment Trials for training TensorPack Faster-RCNN/Mask-RCNN and AWS Samples Mask R-CNN in Amazon SageMaker using Amazon S3 as data source."

However, once the notebook loads the dataset, the Define SageMaker Data Channels section then mentions we are using EFS for training:

"Define SageMaker Data Channels

Next, we define the train data channel using EFS file-system. To do so, we need to specify the EFS file-system id, which is shown in the output of the command below."

If the notebook is going to utilize EFS, then it needs to change the introduction section, and also needs to include this block launching the prepare-efs.sh script, otherwise the job will fail because it won't be able to find the data.

%%time
!./prepare-efs.sh {s3_bucket}

Open mask-rcnn-experiment-trials.ipynb and compare the introduction with the Define SageMaker Data Channels section. Check prepare-efs.sh and the dataset-loading steps first. Done means the notebook consistently describes EFS and includes the prepare-efs.sh launch block needed before training.

Written by the indexing model from the issue text.