-
Proposal to fine-tune the questions in the new project template: https://github.com/hpsfoundation/tac/blob/main/.github/ISSUE_TEMPLATE/new-project-proposal.md
Moved out from #2
I think also rel…
-
### 🚀 The feature, motivation and pitch
I'm struggling to figure out how to extend our test suite to include vllm tests. The problem is that by default vllm will take over the whole gpu, which prev…
-
It's come to light that the `LatitudeLongitudeGrid` consumes almost 1 kb of parameter space as an argument on the GPU. This is a problem because at least for some versions of CUDA + GPUs (unsure how m…
-
(This is running on a Nvidia 4090 GPU, with jax '0.4.31')
I had got that is something like the example below. Here, the depth-wise convolution wants the input to be transposed from [batch, sequence…
-
### 🐛 Describe the bug
```python
import torch
print(torch.__version__)
A = torch.tensor([
[1,1,1],
[1,2,2],
[1,2,3]
], dtype=torch.float32)
l, u = torch.linalg…
-
### 🚀 The feature, motivation and pitch
MI50 is like 2080ti ,but so much cheaper(1/4), and with 16GB memory.
But when I tried to compile it in MI50 machine, I got this:
[ 83%] Building HIP obj…
-
Lỗi khi chạy inference cả model detect và rec cùng 1 lúc bằng code này:
!python tools/infer/predict_system.py \
--image_dir="./train_data/vietnamese/test_image/im1500.jpg" \
--det_model_di…
-
Hey,
I have seen the previous issues. Based on that I tracked down the approximate lane where the pipeline is struck which is the setup function where it failed to load the model. The training is n…
-
### Problem Description
When use api trace on vllm inference, rocprof get less kernel dispatch records than rocprof_v2, which result tend to be correct?
Possible reasons for the mismatch between ker…
-
**Your question**
Ask a clear and concise question about Flux.
There is torch.Size([5120, 1024]) x torch.Size([8192, 1024]) gemm_rs op in my project,fp16.I made a benchmark on A100:
torch.Size(…