AITemplate icon indicating copy to clipboard operation
AITemplate copied to clipboard

AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU) and MatrixCore (AMD GPU) inference.

Results 178 AITemplate issues
Sort by recently updated
recently updated
newest added

Using a DDIM scheduler, I get different results: ![CleanShot 2022-11-27 at 12 30 22](https://user-images.githubusercontent.com/66225/204158153-411c5dac-99d0-4cbb-921e-cd006af744c5.png) Is anyone else getting predictable results with the same seed/scheduler? It makes me think there's a...

Currently there are StableDiffusion examples here: - https://github.com/facebookincubator/AITemplate/tree/main/examples/05_stable_diffusion But it would be awesome if there was also examples of using this for Dreambooth: - https://dreambooth.github.io/ - https://github.com/huggingface/diffusers/tree/main/examples/dreambooth ## See also:...

Hi! I'm trying to compile the Stable Diffusion demo on a RTX 2080, my specs: Ubuntu 22.04 CUDA 11.7 `2022-11-21 18:41:36,544 INFO compiled the final .so file elapsed time: 0:02:44.633613...

Hi! there were 2 issues, about disabling T4 and supporting A10G: https://github.com/facebookincubator/AITemplate/pull/18 https://github.com/facebookincubator/AITemplate/issues/14 there was also a commit with changes in detect_target.py https://github.com/facebookincubator/AITemplate/pull/18 , however 0.1.1 reverted the changes -...

Is branching supported currently? Or does it purely rely on tracing to generate computational graph?

Hello, I have tried to execute the VIT example from the repo in the provided Docker image. I encounter compilation errors during the execution of compile_vit(). Here is a part...

Hi, I'm playing around with ConvTranspose2dBias and I can seem to get it to work. I'm having a few issues with it: 1. The infer_shape() assertion in conv2d seems to...

After looking at the examples of resnet, I didnt find a way to create batchnorm layer using `aitemplate.nn`. Should I have to asusme the nn.conv2d is a fusion of torch.nn.conv2d...

File "\aitemplate\examples\07_how_to_run_pt_model\how_to_run_pt_model.py", line 131, in verify_simple_model() File "\aitemplate\examples\07_how_to_run_pt_model\how_to_run_pt_model.py", line 97, in verify_simple_model with compile_model( File "\AppData\Local\Programs\Python\Python310\lib\site-packages\aitemplate\compiler\compiler.py", line 200, in compile_model compiler.transform.profile( File "\AppData\Local\Programs\Python\Python310\lib\site-packages\aitemplate\compiler\transform\profile.py", line 88, in profile compile_engine.make_profilers(generated_profilers, profiler_dir) File...

Hello, When I run the examples in the ROCm docker, I always get the following error: Traceback (most recent call last): File "examples/07_how_to_run_pt_model/how_to_run_pt_model.py", line 131, in verify_simple_model() File "examples/07_how_to_run_pt_model/how_to_run_pt_model.py", line...