Error while exporting to INT8

[AmeenAli]

Hello

I am using the following detectron2 model :

MODEL:
META_ARCHITECTURE: "GeneralizedRCNN"
BACKBONE:
NAME: "build_resnet_fpn_backbone"
RESNETS:
OUT_FEATURES: ["res2", "res3", "res4", "res5"]
DEPTH: 101
FPN:
IN_FEATURES: ["res2", "res3", "res4", "res5"]
ANCHOR_GENERATOR:
SIZES: [[32], [64], [128], [256], [512]] # One size for each in feature map
ASPECT_RATIOS: [[0.5, 1.0, 2.0]] # Three aspect ratios (same for all in feature maps)
RPN:
IN_FEATURES: ["p2", "p3", "p4", "p5", "p6"]
PRE_NMS_TOPK_TRAIN: 2000 # Per FPN level
PRE_NMS_TOPK_TEST: 1000 # Per FPN level
POST_NMS_TOPK_TRAIN: 1000
POST_NMS_TOPK_TEST: 1000

However, once I try to export it to INT8 by following the demo notebook, I get the following error:
NotImplementedError: Could not run 'quantized::conv2d_prepack' with arguments from the 'QuantizedCUDA' backend. This could be because the operator doesn't exist for this backend, or was omitted during the selective/custom build process (if using custom build). If you are a Facebook employee using PyTorch on mobile, please visit https://fburl.com/ptmfixes for possible resolutions. 'quantized::conv2d_prepack' is only available for these backends: [QuantizedCPU, BackendSelect, Python, Functionalize, Named, Conjugate, Negative, ZeroTensor, ADInplaceOrView, AutogradOther, AutogradCPU, AutogradCUDA, AutogradXLA, AutogradMPS, AutogradXPU, AutogradHPU, AutogradLazy, Tracer, AutocastCPU, AutocastCUDA, Batched, VmapMode, PythonTLSSnapshot].

Any idea what's wrong?

Thanks,

[wat3rBro]

@AmeenAli the quantization for original Detectron2 model is not fully supported.