Hong-Rong Hsu
Hong-Rong Hsu
As the comment mentioned: "Delete the copy constructor and use Clone explicitly" const auto& data = tensor.Clone().data();
use GetMutable or blob_.Get().Clone()
I am building with `riscv64-unknown-linux-gnu-clang`. (The toolchain can be built from the script [here](https://chromium.googlesource.com/libyuv/libyuv/+/refs/heads/main/riscv_script/prepare_toolchain_qemu.sh)) I found it can build pass by removing `-std=c99`. It might be a common issue in...
@fbarchard Could you tell me which risc-v toolchain you are using? Did you encounter the same issue?
Looks like there are some RVV kernels are implemented: https://github.com/google/XNNPACK/commit/aac51aad45a24f1a1acec0a521ac63b6892028b6#diff-3f710269879983f53830fec90d0963ef9da0382cd4eed8bcaf699e8f4101560b What's your(google) plan on it? Will you keep finishing all of the RVV kernels?
@fbarchard Are you working on it?
@mcharleb The mobileBERT model in https://github.com/mlcommons/mobile_models points to https://github.com/fatihcakirs/mobile_models. They are the same.
Hi @benvanik , do you know any follow-ups or plan of the multiple devices support feature?
@Hacker1337 yes I got the same error with output to StableHLO. Try to roll back the transformers version back to 4.40.0 works. ``` pip install transformers==4.40.0 ``` But it just...
Thanks for the information. I've tried `pip install https://download.pytorch.org/whl/nightly/cpu/torch-2.4.0.dev20240612%2Bcpu-cp310-cp310-linux_x86_64.whl` on my ubuntu machine and it works.