Alexander Dokuchaev

@basioli-k Thanks to provide code, this problem is actual for v270. But issues is not reproduced on `develop` and `release_v280` branches. Try to install NCCF from `develop` branch by `pip...

The link checker failed due to links that will only exist in PyPiPublishing.md after the merge. ``` File: docs/PyPiPublishing.md:45 • Link: https://github.com/openvinotoolkit/nncf/blob/develop/src/nncf/experimental/torch/sparsify_activations/ActivationSparsity.md • Error: 404: Not Found File: docs/PyPiPublishing.md:56 •...

@ljaljushkin https://github.com/openvinotoolkit/nncf/actions/runs/15973365309 https://github.com/openvinotoolkit/nncf/actions/runs/15973378051 https://github.com/openvinotoolkit/nncf/actions/runs/15973363013

Example when "forward" in `model.__dict__` raise exception ```python from transformers import AutoModelForImageClassification import torch model = AutoModelForImageClassification.from_pretrained("microsoft/resnet-18").cuda() inputs = torch.rand(10, 3, 224, 224).cuda() # Set forward like https://github.com/huggingface/optimum/blob/v1.24.0-release/optimum/exporters/onnx/model_patcher.py#L292 setattr(model, "forward",...

@IlyasMoutawwakil ``` 1. load model 2. export 3. train as DataParallel

Please add new tests to https://github.com/openvinotoolkit/nncf/blob/develop/tests/cross_fw/examples/.test_durations

Looks like it actual for PT and ONNX too. And please check may be possible add some tests to template.

HI @zbnlala Could you please provide a full script to reproduce the issue?

@zbnlala unfortunately this issue is not reproduced. Could you check that issue is not reproduced without nncf?

Please add unit tests for constant folding and docstrings.