gen2-maskrcnn-resnet50 error

[laitathei]

I have followed https://github.com/openvinotoolkit/openvino_contrib/tree/master/modules/mo_pytorch guide line to convert model. it seems no problem with this guideline. Alexnet model can successfully convert but maskrcnn_resnet50_fpn cannot be able to convert.

The following code is what i have done:

import torchvision.models as models
from torchvision import models
import torchvision

# Create model
# model = models.alexnet(pretrained=True)
# model = models.resnet50(True) 
#model = models.alexnet(weights=models.AlexNet_Weights.DEFAULT)
model = torchvision.models.detection.maskrcnn_resnet50_fpn(weights=torchvision.models.detection.MaskRCNN_ResNet50_FPN_Weights.DEFAULT)
# Convert to OpenVINO IR
import mo_pytorch
mo_pytorch.convert(model, input_shape=[1, 3, 227, 227], model_name='maskrcnn_resnet50_fpn')

Then, it generate following error

Common parameters:
    - Path to the Input Model:  MaskRCNN(
  (transform): GeneralizedRCNNTransform(
      Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
      Resize(min_size=(800,), max_size=1333, mode='bilinear')
  )
  (backbone): BackboneWithFPN(
    (body): IntermediateLayerGetter(
      (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
      (bn1): FrozenBatchNorm2d(64, eps=0.0)
      (relu): ReLU(inplace=True)
      (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
      (layer1): Sequential(
        (0): Bottleneck(
          (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(64, eps=0.0)
          (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(64, eps=0.0)
          (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(256, eps=0.0)
          (relu): ReLU(inplace=True)
          (downsample): Sequential(
            (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
            (1): FrozenBatchNorm2d(256, eps=0.0)
          )
        )
        (1): Bottleneck(
          (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(64, eps=0.0)
          (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(64, eps=0.0)
          (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(256, eps=0.0)
          (relu): ReLU(inplace=True)
        )
        (2): Bottleneck(
          (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(64, eps=0.0)
          (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(64, eps=0.0)
          (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(256, eps=0.0)
          (relu): ReLU(inplace=True)
        )
      )
      (layer2): Sequential(
        (0): Bottleneck(
          (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(128, eps=0.0)
          (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(128, eps=0.0)
          (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(512, eps=0.0)
          (relu): ReLU(inplace=True)
          (downsample): Sequential(
            (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
            (1): FrozenBatchNorm2d(512, eps=0.0)
          )
        )
        (1): Bottleneck(
          (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(128, eps=0.0)
          (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(128, eps=0.0)
          (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(512, eps=0.0)
          (relu): ReLU(inplace=True)
        )
        (2): Bottleneck(
          (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(128, eps=0.0)
          (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(128, eps=0.0)
          (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(512, eps=0.0)
          (relu): ReLU(inplace=True)
        )
        (3): Bottleneck(
          (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(128, eps=0.0)
          (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(128, eps=0.0)
          (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(512, eps=0.0)
          (relu): ReLU(inplace=True)
        )
      )
      (layer3): Sequential(
        (0): Bottleneck(
          (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(256, eps=0.0)
          (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(256, eps=0.0)
          (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(1024, eps=0.0)
          (relu): ReLU(inplace=True)
          (downsample): Sequential(
            (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
            (1): FrozenBatchNorm2d(1024, eps=0.0)
          )
        )
        (1): Bottleneck(
          (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(256, eps=0.0)
          (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(256, eps=0.0)
          (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(1024, eps=0.0)
          (relu): ReLU(inplace=True)
        )
        (2): Bottleneck(
          (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(256, eps=0.0)
          (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(256, eps=0.0)
          (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(1024, eps=0.0)
          (relu): ReLU(inplace=True)
        )
        (3): Bottleneck(
          (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(256, eps=0.0)
          (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(256, eps=0.0)
          (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(1024, eps=0.0)
          (relu): ReLU(inplace=True)
        )
        (4): Bottleneck(
          (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(256, eps=0.0)
          (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(256, eps=0.0)
          (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(1024, eps=0.0)
          (relu): ReLU(inplace=True)
        )
        (5): Bottleneck(
          (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(256, eps=0.0)
          (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(256, eps=0.0)
          (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(1024, eps=0.0)
          (relu): ReLU(inplace=True)
        )
      )
      (layer4): Sequential(
        (0): Bottleneck(
          (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(512, eps=0.0)
          (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(512, eps=0.0)
          (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(2048, eps=0.0)
          (relu): ReLU(inplace=True)
          (downsample): Sequential(
            (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
            (1): FrozenBatchNorm2d(2048, eps=0.0)
          )
        )
        (1): Bottleneck(
          (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(512, eps=0.0)
          (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(512, eps=0.0)
          (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(2048, eps=0.0)
          (relu): ReLU(inplace=True)
        )
        (2): Bottleneck(
          (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn1): FrozenBatchNorm2d(512, eps=0.0)
          (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): FrozenBatchNorm2d(512, eps=0.0)
          (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn3): FrozenBatchNorm2d(2048, eps=0.0)
          (relu): ReLU(inplace=True)
        )
      )
    )
    (fpn): FeaturePyramidNetwork(
      (inner_blocks): ModuleList(
        (0): Conv2dNormActivation(
          (0): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1))
        )
        (1): Conv2dNormActivation(
          (0): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1))
        )
        (2): Conv2dNormActivation(
          (0): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1))
        )
        (3): Conv2dNormActivation(
          (0): Conv2d(2048, 256, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (layer_blocks): ModuleList(
        (0): Conv2dNormActivation(
          (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
        (1): Conv2dNormActivation(
          (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
        (2): Conv2dNormActivation(
          (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
        (3): Conv2dNormActivation(
          (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        )
      )
      (extra_blocks): LastLevelMaxPool()
    )
  )
  (rpn): RegionProposalNetwork(
    (anchor_generator): AnchorGenerator()
    (head): RPNHead(
      (conv): Sequential(
        (0): Conv2dNormActivation(
          (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
          (1): ReLU(inplace=True)
        )
      )
      (cls_logits): Conv2d(256, 3, kernel_size=(1, 1), stride=(1, 1))
      (bbox_pred): Conv2d(256, 12, kernel_size=(1, 1), stride=(1, 1))
    )
  )
  (roi_heads): RoIHeads(
    (box_roi_pool): MultiScaleRoIAlign(featmap_names=['0', '1', '2', '3'], output_size=(7, 7), sampling_ratio=2)
    (box_head): TwoMLPHead(
      (fc6): Linear(in_features=12544, out_features=1024, bias=True)
      (fc7): Linear(in_features=1024, out_features=1024, bias=True)
    )
    (box_predictor): FastRCNNPredictor(
      (cls_score): Linear(in_features=1024, out_features=91, bias=True)
      (bbox_pred): Linear(in_features=1024, out_features=364, bias=True)
    )
    (mask_roi_pool): MultiScaleRoIAlign(featmap_names=['0', '1', '2', '3'], output_size=(14, 14), sampling_ratio=2)
    (mask_head): MaskRCNNHeads(
      (0): Conv2dNormActivation(
        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        (1): ReLU(inplace=True)
      )
      (1): Conv2dNormActivation(
        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        (1): ReLU(inplace=True)
      )
      (2): Conv2dNormActivation(
        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        (1): ReLU(inplace=True)
      )
      (3): Conv2dNormActivation(
        (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        (1): ReLU(inplace=True)
      )
    )
    (mask_predictor): MaskRCNNPredictor(
      (conv5_mask): ConvTranspose2d(256, 256, kernel_size=(2, 2), stride=(2, 2))
      (relu): ReLU(inplace=True)
      (mask_fcn_logits): Conv2d(256, 91, kernel_size=(1, 1), stride=(1, 1))
    )
  )
)
    - Path for generated IR:    /home/teddylai/Desktop/Pytorch_Mask_RCNN/.
    - IR output name:   maskrcnn_resnet50_fpn
    - Log level:    ERROR
    - Batch:    Not specified, inherited from the model
    - Input layers:     Not specified, inherited from the model
    - Output layers:    Not specified, inherited from the model
    - Input shapes:     [1, 3, 227, 227]
    - Source layout:    Not specified
    - Target layout:    Not specified
    - Layout:   Not specified
    - Mean values:  Not specified
    - Scale values:     Not specified
    - Scale factor:     Not specified
    - Precision of IR:  FP32
    - Enable fusing:    True
    - User transformations:     Not specified
    - Reverse input channels:   False
    - Enable IR generation for fixed input shape:   False
    - Use the transformations config file:  None
Advanced parameters:
    - Force the usage of legacy Frontend of Model Optimizer for model conversion into IR:   False
    - Force the usage of new Frontend of Model Optimizer for model conversion into IR:  False
/home/teddylai/.local/lib/python3.8/site-packages/torch/nn/modules/conv.py:453: UserWarning: Defining your `__torch_function__` as a plain method is deprecated and will be an error in future, please define it as a classmethod. (Triggered internally at  ../torch/csrc/utils/python_arg_parser.cpp:295.)
  return F.conv2d(input, weight, bias, self.stride,
[ ERROR ]  -------------------------------------------------
[ ERROR ]  ----------------- INTERNAL ERROR ----------------
[ ERROR ]  Unexpected exception happened.
[ ERROR ]  Please contact Model Optimizer developers and forward the following information:
[ ERROR ]  Exception occurred during running replacer "REPLACEMENT_ID (<class 'mo_extensions.load.pytorch.loader.PyTorchLoader'>)": __init__() got an unexpected keyword argument 'antialias'
[ ERROR ]  Traceback (most recent call last):
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/utils/class_registration.py", line 278, in apply_transform
    for_graph_and_each_sub_graph_recursively(graph, replacer.find_and_replace_pattern)
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/middle/pattern_match.py", line 46, in for_graph_and_each_sub_graph_recursively
    func(graph)
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/load/loader.py", line 14, in find_and_replace_pattern
    self.load(graph)
  File "/home/teddylai/Desktop/Pytorch_Mask_RCNN/openvino_contrib/modules/mo_pytorch/mo_extensions/load/pytorch/loader.py", line 96, in load
    model(inputs['input'])
  File "/home/teddylai/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/teddylai/Desktop/Pytorch_Mask_RCNN/openvino_contrib/modules/mo_pytorch/mo_extensions/load/pytorch/mask_rcnn/mask_rcnn.py", line 188, in <lambda>
    model.forward = lambda *args: model_forward(model, *args)
  File "/home/teddylai/Desktop/Pytorch_Mask_RCNN/openvino_contrib/modules/mo_pytorch/mo_extensions/load/pytorch/mask_rcnn/mask_rcnn.py", line 176, in model_forward
    features = self.backbone(images.tensors)
  File "/home/teddylai/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/teddylai/.local/lib/python3.8/site-packages/torchvision/models/detection/backbone_utils.py", line 58, in forward
    x = self.fpn(x)
  File "/home/teddylai/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/teddylai/.local/lib/python3.8/site-packages/torchvision/ops/feature_pyramid_network.py", line 194, in forward
    inner_top_down = F.interpolate(last_inner, size=feat_shape, mode="nearest")
  File "/home/teddylai/.local/lib/python3.8/site-packages/torch/nn/functional.py", line 3819, in interpolate
    return handle_torch_function(
  File "/home/teddylai/.local/lib/python3.8/site-packages/torch/overrides.py", line 1498, in handle_torch_function
    result = torch_func_method(public_api, types, args, kwargs)
  File "/home/teddylai/Desktop/Pytorch_Mask_RCNN/openvino_contrib/modules/mo_pytorch/mo_extensions/load/pytorch/hooks.py", line 463, in __torch_function__
    return HANDLED_FUNCTIONS[func](*args, **kwargs)
  File "/home/teddylai/Desktop/Pytorch_Mask_RCNN/openvino_contrib/modules/mo_pytorch/mo_extensions/load/pytorch/hooks.py", line 780, in function_hook
    return forward_hook(Upsample(*args, **kwargs), (input,))
TypeError: __init__() got an unexpected keyword argument 'antialias'

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/main.py", line 533, in main
    ret_code = driver(argv)
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/main.py", line 489, in driver
    graph, ngraph_function = prepare_ir(argv)
  File "/home/teddylai/Desktop/Pytorch_Mask_RCNN/openvino_contrib/modules/mo_pytorch/mo_pytorch.py", line 99, in _prepare_ir
    graph = unified_pipeline(argv)
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/pipeline/unified.py", line 13, in unified_pipeline
    class_registration.apply_replacements(graph, [
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/utils/class_registration.py", line 328, in apply_replacements
    apply_replacements_list(graph, replacers_order)
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/utils/class_registration.py", line 314, in apply_replacements_list
    apply_transform(
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/utils/logger.py", line 112, in wrapper
    function(*args, **kwargs)
  File "/home/teddylai/.local/lib/python3.8/site-packages/openvino/tools/mo/utils/class_registration.py", line 302, in apply_transform
    raise Exception('Exception occurred during running replacer "{} ({})": {}'.format(
Exception: Exception occurred during running replacer "REPLACEMENT_ID (<class 'mo_extensions.load.pytorch.loader.PyTorchLoader'>)": __init__() got an unexpected keyword argument 'antialias'

[ ERROR ]  ---------------- END OF BUG REPORT --------------
[ ERROR ]  -------------------------------------------------
Traceback (most recent call last):
  File "export_onnx2.py", line 12, in <module>
    mo_pytorch.convert(model, input_shape=[1, 3, 227, 227], model_name='maskrcnn_resnet50_fpn')
  File "/home/teddylai/Desktop/Pytorch_Mask_RCNN/openvino_contrib/modules/mo_pytorch/mo_pytorch.py", line 129, in convert
    raise Exception('model conversion failed')
Exception: model conversion failed

It is good for providing tutorial of https://github.com/luxonis/depthai-experiments/tree/master/gen2-maskrcnn-resnet50 for converting the model from pytorch to openvino. Thank you.

[Erol444]

Hi @laitathei , we apologize for the delay, I just pinged the ML team for assistance here.

[tersekmatija]

Hey @laitathei , we are investigating. In the meantime could you try the code from this comment in #353. It was tested with:

torch==1.9.1
torchvision==0.10.0
openvino-dev==2021.4.2

[klemen1999]

Hi @laitathei , based on @tersekmatija 's comment and by investigating mo_pytorch, we've found a working solution for exporting Mask R-CNN model with openvino==2021.4.2. Here is a notebook that produces a blob which you can use in the mentioned example.

[laitathei]

@klemen1999 thank you for your helping. The notebook is exactly what i want. It can convert the mask rcnn model to blob format very well !