This repository contains example notebooks and documentation for working with Verily Workbench.
Verily Workbench is a new platform developed by Verily to connect the multimodal data ecosystem and meet the needs of commercial or industry-led research organizations. Building on Terra’s academic roots, Verily Workbench is an integrated platform that provides a secure foundation to discover, govern, and unify multimodal data in a collaborative environment to accelerate discoveries.
- Explore: discover multimodal data sets available in Verily Workbench’s data ecosystem.
- Unify: connect all data types, from both internal and external sources, into a single platform under one governance framework.
- Analyze: unlock insights faster with familiar analysis tools.
- Collaborate: share best practices and publish reproducible research.
- Scale: maximize the power of any cloud with security, compliance, and support from industry standards.
- Visit Verily Workbench
- Add your SSH key to GitHub.
- Create a personal workspace where you can run examples, including the ones in this repo. Name it something like "Personal workspace {your email address} {YYYYMMDD}".
- [Optional] If you want Verily Workbench to clone this repo automatically, add it as a reference in your
workspace. Otherwise, you
can clone it manually:
git clone https://github.com/verily-src/workbench-examples.git
- Create a cloud environment for yourself. Name it something like "{your email address}-{YYYYMMDD}".
- Be sure to run the notebook workspace_setup.ipynb first before running any other notebook files. It sets up some buckets and a BigQuery dataset that all the other notebooks refer to by reference and expect to exist. You only need to run this notebook once per workspace, though it's fine to rerun it.
- Run the notebook cloud_env_setup.ipynb for each cloud environment that you create. It does some useful git and virtual environment configuration. You may find it useful to add your own environment config to this notebook as well.