cutlass
shark-samples
Our great sponsors
cutlass | shark-samples | |
---|---|---|
16 | 1 | |
4,496 | 15 | |
5.5% | - | |
8.8 | 0.0 | |
4 days ago | about 2 years ago | |
C++ | Python | |
GNU General Public License v3.0 or later | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
cutlass
-
Optimization Techniques for GPU Programming [pdf]
I would recommend the course from Oxford (https://people.maths.ox.ac.uk/gilesm/cuda/). Also explore the tutorial section of cutlass (https://github.com/NVIDIA/cutlass/blob/main/media/docs/cute/...) if you want to learn more about high performance gemm.
-
Want to understand INT8 better
The latter (and I guess you were asking about this one) is designed to accelerate NN inference in reduced precision. It is possible to use Tensor Cores for you own purposes, mainly through CUTLASS. But because Tensor Cores are designed to execute matrix multiplications, it can be hard to adapt your problem to them. The performance with them is insane (IIRC 32x the performance of the INT32 pipeline), but only for matrix multiplication…
-
How do I deal with tensor core and cuda core with different precision?
If you want to learn about controlling Tensor Cores, the main way is through the CUTLASS library, that wraps the complexity of Tensor Cores into higher level abstractions. You can also look for mma/wmma instructions in the PTX specification, or for the WMMA API in CUDA.
-
AI’s compute fragmentation: what matrix multiplication teaches us
> we used tensor cores and managed to get back fp32 accuracy with 3 rounds of the things
Hey are you referring to 3xTF32 (https://github.com/NVIDIA/cutlass/tree/master/examples/28_am...)? IMO this is a perfect example where proper abstraction could save engineers non-trivial amount of time - imagine a compiler stack which allows 3xTF32 as a normal dtype and subsequent analysis compatible with this special dtype :-)
- With LLVM and MLIR, is manual cuda optimizing still important?
- CUTLASS 3.0 is now available
-
How to Optimize a CUDA Matmul Kernel for CuBLAS-Like Performance: A Worklog
This is a great post for people who are new to optimizing GPU code.
It is interesting to see that the author got this far without interchanging the innermost loop over k to the outermost loop, as is done in CUTLASS (https://github.com/NVIDIA/cutlass).
As you can see in this blog post the code ends up with a lot of compile-time constants (e.g. BLOCKSIZE, BM, BN, BK, TM, TN) one way to optimize this code further is to use an auto-tuner to find the optimal value for all of these parameters for your GPU and problem size, for example Kernel Tuner (https://github.com/KernelTuner/kernel_tuner)
-
pytorch example to actually see anything near 83 TFLOP/s on a RTX 4090?
Some examples here have a benchmark: https://github.com/NVIDIA/cutlass/blob/master/examples/24_gemm_grouped/gemm_grouped.cu
-
Create a bare CMake for Nvidia CUTLASS
I would like to make a minimum CMakeLists to use the CUDA CUTLASS library in another project. The build system is CMake, however I have little experience with CMake.
-
[D] What are some good resources to learn CUDA programming?
If you already know some C++, the Nvidia devblog is a great resource. Going further, Cub and Cutlass provide examples of efficient implementations for key operations at all hardware levels. Finally, this is more anecdotal but I always start my lectures on Cuda programming with the pictures in this doc page, to provide some intuition on the different memory layers that you can leverage to speed up a program. In any case, good luck :-)
shark-samples
-
PyTorch on Apple M1 Faster Than TensorFlow-Metal
I updated the blog with the reference. Basically it crashes to compile the model with https://github.com/NodLabs/shark-samples/blob/main/examples/.... The coremltools converter is very version specific (like all vendor conversion kits) and still on a version of TF I couldn't get on conda. Also it doesn't allow for training and only FP16 for inference with ANE. All our tests were with FP32.
//part of nod.ai/shark team.
What are some alternatives?
TensorRT - PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT
iree - A retargetable MLIR-based machine learning compiler and runtime toolkit.
mmperf - MatMul Performance Benchmarks for a Single CPU Core comparing both hand engineered and codegen kernels.
GPU-Puzzles - Solve puzzles. Learn CUDA.
triton - Development repository for the Triton language and compiler
Chess_BinaryNeuralNetwork - Training and Code Emitting Library for Binary Neural Networks
Open3D - Open3D: A Modern Library for 3D Data Processing
maxas - Assembler for NVIDIA Maxwell architecture
flopth - A simple program to calculate and visualize the FLOPs and Parameters of Pytorch models, with handy CLI and easy-to-use Python API.
kernel_tuner_tutorial - A hands-on introduction to tuning GPU kernels using Kernel Tuner https://github.com/KernelTuner/kernel_tuner/