-
- Downloads
[Torch] Experimental support for FX-quantized models (#10091)
* works on resnet18 and deeplabv3 * yolo5 conversion worked * fixed sigmoid * [Torch] Support clamp_min, clamp_max * fixed clamp_min * fixed quantize for 1 dim input * cleanup * improve inline_qparam impl * add clamp_min/max test * add fx quant test * cleanup * skip build in testing * black * improve clamp conversion * leave TODO on inf handling
Showing
- python/tvm/relay/frontend/pytorch.py 46 additions, 19 deletionspython/tvm/relay/frontend/pytorch.py
- python/tvm/relay/frontend/pytorch_utils.py 6 additions, 0 deletionspython/tvm/relay/frontend/pytorch_utils.py
- python/tvm/relay/frontend/qnn_torch.py 67 additions, 17 deletionspython/tvm/relay/frontend/qnn_torch.py
- tests/python/frontend/pytorch/test_forward.py 4 additions, 0 deletionstests/python/frontend/pytorch/test_forward.py
- tests/python/frontend/pytorch/test_fx_quant.py 85 additions, 0 deletionstests/python/frontend/pytorch/test_fx_quant.py
Loading
Please register or sign in to comment