Wang, Kai Lawrence comments

Results 15 comments of


                                            Wang, Kai Lawrence

Using XPU training actually increases the training time and sharply reduces the accuracy.

Hi @SoldierWz, thanks for reporting this issue. Could you please help provide the env info via the following commands: ``` wget https://raw.githubusercontent.com/intel/intel-extension-for-pytorch/master/scripts/collect_env.py # For security purposes, please check the contents...

Using XPU training actually increases the training time and sharply reduces the accuracy.

@SoldierWz XPU seems not be detected by IPEX in your environment. IPEX 2.1.10+xpu works with oneAPI 2024.0, please help downgrade the dpcpp and mkl version via `sudo apt install -y...

Using XPU training actually increases the training time and sharply reduces the accuracy.

@SoldierWz @Shr1ftyy For the undefined symbol issue caused by the mismatch between IPEX 2.1.10+xpu and oneAPI 2024.1, we recently released IPEX 2.1.20+xpu, which is compatible with oneAPI base toolkit 2024.1,...

Using XPU training actually increases the training time and sharply reduces the accuracy.

@SoldierWz BTW, the dataset you shared seems not successfully uploaded. Please help check.

Using XPU training actually increases the training time and sharply reduces the accuracy.

There is no response on this issue for over 1 month. Close the issue now. If you still have issues, feel free to reopen it. Thanks.

No visible XPU devices when running intel-extension-for-tensorflow/tree/main/examples/train_maskrcnn on PVC

@ch0801 `E external/intel_xla/xla/stream_executor/sycl/sycl_gpu_runtime.cc:178] Can not found any devices.` From the output, the XPU devices are not detected via the framework. Could you please run the following commands to collect the...

No visible XPU devices when running intel-extension-for-tensorflow/tree/main/examples/train_maskrcnn on PVC

@ch0801 Via your output of env check tool, the XPU are still not detected by the framework during runtime. ITEX v2.15.0.0 typically works with oneAPI base Toolkit 2024.1 while you...

No visible XPU devices when running intel-extension-for-tensorflow/tree/main/examples/train_maskrcnn on PVC

@Millionarc Please provide your output for running the `env_check.py`.

No visible XPU devices when running intel-extension-for-tensorflow/tree/main/examples/train_maskrcnn on PVC

Worked with @ch0801's team and confirmed the issue has gone now. XPU can be detected and the workloads are able to execute on the GPUs. Close it.

Running IPEX samples - environment setup clarification

@js333031 Thanks for reporting the issue. Yes, the ipex will be built from source anyway in the main branch, while using the release branch such as `2.3.100+cpu` will have the...