search

intel / torch-xpu-ops

Apache License 2.0
30 stars 21 forks source link

Add quantized_maxpool_2d for xpu #1049

Open gaopengff opened 2 weeks ago

gaopengff commented 2 weeks ago

Now we only support datatype of uint8(Byte). Referring the stock pytorch cpu implementation at code. Waiting https://github.com/intel/torch-xpu-ops/pull/921 to be merged.

gaopengff commented 2 weeks ago

The specific unit test cases have passed in latest CI.

2024-11-07T09:10:50.0722836Z quantization/core/test_quantized_op_xpu.py::TestQuantizedOpsXPU::test_max_pool2d_nhwc_xpu PASSED [ 48%]
2024-11-07T09:10:50.1551494Z quantization/core/test_quantized_op_xpu.py::TestQuantizedOpsXPU::test_max_pool2d_pt2e_xpu PASSED [ 50%]
2024-11-07T09:10:50.5847030Z quantization/core/test_quantized_op_xpu.py::TestQuantizedOpsXPU::test_max_pool2d_xpu PASSED [ 51%]
ZhiweiYan-96 commented 3 days ago

Please fix ci error