You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
They also have a tenforflow landsat example which is just doing some tensor math to convert satellite data to a JPEG. It would be quite straight forward to convert to a cupy example. The input dataset here is a list of filenames, each one containing a single landsat image.
For our page I expect we would want:
Build a container from the RAPIDS base and add Apache Beam
Write a small example workload that uses RAPIDS
Execute this on Dataflow
The text was updated successfully, but these errors were encountered:
Add documentation page to the Google Cloud section on using RAPIDS within Dataflow.
Google have a minimal pytorch example in their documentation. It builds a container on the pytorch one and brings in the apache beam dependencies, so we could do the same with RAPIDS. It then has a small code example that invokes
pytorch
.They also have a tenforflow landsat example which is just doing some tensor math to convert satellite data to a JPEG. It would be quite straight forward to convert to a cupy example. The input dataset here is a list of filenames, each one containing a single landsat image.
For our page I expect we would want:
The text was updated successfully, but these errors were encountered: