Tutorials/Accelerated Python: Fixes and improvements to memory spaces, asynchrony, and kernel authoring notebooks by brycelelbach · Pull Request #148 · NVIDIA/accelerated-computing-hub

Tutorials/Accelerated Python/Kernel Authoring: Improve book histogram plot formatting, add title and dataset size display.

brycelelbach

committed

Tutorials/Accelerated Python/Kernel Authoring: Add configurable correctness check to copy kernel launch function.

brycelelbach

committed

Tutorials/Accelerated Python/Kernel Authoring: Display book histogram dataset size in megabytes instead of bytes.

brycelelbach

committed

Tutorials/Accelerated Python/Kernel Authoring: Add correctness check cell before profiling copy_blocked kernel.

brycelelbach

committed

Tutorials/Accelerated Python/Power Iteration: Fix typos in example usage of cpyx.profiler.profile and NVTX.

brycelelbach

committed

Tutorials/Accelerated Python/Kernel Authoring: Apply title, TOC, section header, and text updates to solution notebooks.

brycelelbach

committed

Tutorials/Accelerated Python/Kernel Authoring: Add output mode to copy kernel scripts to print problem size and dtype.

brycelelbach

committed

Tutorials/Accelerated Python/Memory Spaces: Re-add checkpoint I/O to power iteration.

brycelelbach

committed

Tutorials/Accelerated Python/Memory Spaces: Remove unnecessary isinstance check before cp.asarray().

brycelelbach

committed

Tutorials/Accelerated Python/Memory Spaces: Rename estimate_device_exercise and generate_device_exercise back to estimate_device and generate_device.

brycelelbach

committed

Tutorials/Accelerated Python/Asynchrony: Limit warmup calls to 1 step to avoid unnecessary computation.

brycelelbach

committed

Tutorials/Accelerated Python/Memory Spaces: Revert benchmarking to use the same input matrix for host and device.

brycelelbach

committed

Tutorials/Accelerated Python: Unify benchmark reporting to use milliseconds with mean ± relative stdev format.

brycelelbach

committed

Tutorials/Accelerated Python/Memory Spaces: Separate eigvals into its own cell, fix capitalization, restore eigvals timing.

brycelelbach

committed

Tutorials/Accelerated Python/Asynchrony: Fix compute step NVTX annotations with accurate step ranges and per-step regions.

brycelelbach

committed

Tutorials/Accelerated Python/Memory Spaces: Remove inline benchmarking from sections 3 and 4.

brycelelbach

committed

Tutorials/Accelerated Python/Memory Spaces: Remove outdated note about cupy.linalg.eigvals not being implemented.

brycelelbach

committed

Tutorials/Accelerated Python/Memory Spaces: Fix benchmark to compare CPU wall-clock times for both host and device.

brycelelbach

committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Tutorials/Accelerated Python: Fixes and improvements to memory spaces, asynchrony, and kernel authoring notebooks#148

Tutorials/Accelerated Python: Fixes and improvements to memory spaces, asynchrony, and kernel authoring notebooks#148
brycelelbach wants to merge 20 commits intomainfrom
fix/accelerated-python-memory-space-asynchrony-and-kernel-notebooks

Commits on Mar 3, 2026

Tutorials/Accelerated Python/Kernel Authoring: Improve book histogram plot formatting, add title and dataset size display.

Tutorials/Accelerated Python/Kernel Authoring: Add configurable correctness check to copy kernel launch function.

Tutorials/Accelerated Python/Kernel Authoring: Display book histogram dataset size in megabytes instead of bytes.

Tutorials/Accelerated Python/Kernel Authoring: Add correctness check cell before profiling copy_blocked kernel.

Tutorials/Accelerated Python/Power Iteration: Fix typos in example usage of cpyx.profiler.profile and NVTX.

Tutorials/Accelerated Python/Kernel Authoring: Apply title, TOC, section header, and text updates to solution notebooks.

Tutorials/Accelerated Python/Kernel Authoring: Add output mode to copy kernel scripts to print problem size and dtype.

Tutorials/Accelerated Python/Memory Spaces: Re-add checkpoint I/O to power iteration.

Tutorials/Accelerated Python/Memory Spaces: Remove unnecessary isinstance check before cp.asarray().

Tutorials/Accelerated Python/Memory Spaces: Rename estimate_device_exercise and generate_device_exercise back to estimate_device and generate_device.

Tutorials/Accelerated Python/Asynchrony: Limit warmup calls to 1 step to avoid unnecessary computation.

Tutorials/Accelerated Python/Memory Spaces: Revert benchmarking to use the same input matrix for host and device.

Tutorials/Accelerated Python: Unify benchmark reporting to use milliseconds with mean ± relative stdev format.

Tutorials/Accelerated Python/Memory Spaces: Separate eigvals into its own cell, fix capitalization, restore eigvals timing.

Tutorials/Accelerated Python/Asynchrony: Fix compute step NVTX annotations with accurate step ranges and per-step regions.

Tutorials/Accelerated Python/Memory Spaces: Remove inline benchmarking from sections 3 and 4.

Tutorials/Accelerated Python/Memory Spaces: Remove outdated note about cupy.linalg.eigvals not being implemented.

Tutorials/Accelerated Python/Memory Spaces: Fix benchmark to compare CPU wall-clock times for both host and device.

Commits on Mar 6, 2026

Tutorials/Accelerated Python/Kernels: Rename "output" CLI arg to "check" and normalize notebook formatting.

Tutorials/Accelerated Python/Kernels: Fix invalid notebook output metadata in histogram SOLUTION.