MCT fails on torch.nn.functional.layer_norm

[Idan-BenAmi]

Issue Type

Bug

Source

source

MCT Version

1.10.0

OS Platform and Distribution

No response

Python version

No response

Describe the issue

Model that includes layer_norm() raises the following MCT error:
TypeError: layer_norm(): argument 'normalized_shape' (position 2) must be tuple of ints, not Parameter

Expected behaviour

No response

Code to reproduce the issue

model_optimization/tutorials/quick_start/main.py --model_name pvt_v2_b2 --model_library timm --validation_dataset_folder <> --representative_dataset_folder <> --validation_set_limit 100

Log output

File "./venv_np_1_26/lib/python3.11/site-packages/torch/nn/functional.py", line 2515, in layer_norm
    return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: layer_norm(): argument 'normalized_shape' (position 2) must be tuple of ints, not Parameter

[Idan-BenAmi]

Issue was resolved in #923 - we replace functional layer_norm with nn.LayerNorm