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
It would be nice to provide a parallel version of this gallery to illustrate how the cuCIM API can also be used to produce the same result. There are a subset of examples we cannot currently implement due to incomplete coverage, but more than half should be doable with our current API.
One drawback is that the data used in all of these examples tend to be very small. Additionally we have to transfer any results back to the host to plot via matplotlib. Combined these will lead to the examples not really illustrating the benefit of GPU acceleration. I think they are still useful as usage examples, though. We already have the benchmarks folder with scripts there to measure relative performance of operations on CPU vs. GPU.
The text was updated successfully, but these errors were encountered:
scikit-image provides a nice gallery of examples (built via sphinx-gallery).
It would be nice to provide a parallel version of this gallery to illustrate how the cuCIM API can also be used to produce the same result. There are a subset of examples we cannot currently implement due to incomplete coverage, but more than half should be doable with our current API.
One drawback is that the data used in all of these examples tend to be very small. Additionally we have to transfer any results back to the host to plot via matplotlib. Combined these will lead to the examples not really illustrating the benefit of GPU acceleration. I think they are still useful as usage examples, though. We already have the
benchmarks
folder with scripts there to measure relative performance of operations on CPU vs. GPU.The text was updated successfully, but these errors were encountered: