You are planning to build a team data science environment for training machine learning models with datasets exceeding 20 GB. The environment must meet the following requirements:

• Models must be built using either the Caffe2 or Chainer frameworks.
• Data scientists must be able to use the environment to build machine learning pipelines and train models on their personal devices, functioning in both connected and disconnected network states.
• Personal devices must be able to update the machine learning pipelines when a network connection is available.

Which data science environment should you select?