Vinnam Kim
`tests/tensorflow/test_compressed_graph.py .s...s...s......` This PR makes `test_compressed_graph` too long (>4 days), causing our CI to be timeout in the middle of TF precommit testing. Please see `job/NNCF/job/precommit/job/ubuntu20_precommit_tensorflow/73/console`.
Hi @linhdb-2149, This is our quantization forward function for CPU. https://github.com/openvinotoolkit/nncf/blob/13cc15e6560985606abfd55086242d9f7687d6dd/nncf/torch/extensions/src/quantization/cpu/functions_cpu.cpp#L12-L19 You can see our implementation multiplying `s` = $\frac{q_{max}}{scale}$ ($q_{max}$ = `levels - 1` and $scale$ = `input_range`) to...
test this please
`op_exec_context` is an instance of `OperationExecutionContext` which has `self.op_address: OperationAddress`. https://github.com/openvinotoolkit/nncf/blob/9b9c3b36704fe8fe46296573993af12ea16c8294/nncf/torch/dynamic_graph/graph.py#L105-L120 `OperationAddress` includes `Scope`. https://github.com/openvinotoolkit/nncf/blob/9b9c3b36704fe8fe46296573993af12ea16c8294/nncf/torch/dynamic_graph/operation_address.py#L13-L20 Then, `Scope` is used to find `nn.module`. https://github.com/openvinotoolkit/nncf/blob/9b9c3b36704fe8fe46296573993af12ea16c8294/nncf/torch/nncf_network.py#L577-L592
> Guys, looks like we were removed from nncf developers group. Please add me as an reviewer and don't merge this without our review! > > cc: @negved Please request...
> > Guys, looks like we were removed from nncf developers group. Please add me as an reviewer and don't merge this without our review! > > cc: @negved >...
Hi, sorry for the late response. There is currently no roadmap for torch.fx support. We will let you know if there are any changes. Thank you!
[upstream patch](https://github.com/pytorch/pytorch/pull/67202 ) was applied `torch==1.11.0`.
@sungchul2 Please rebase this PR on the latest develop branch and push again. This is because of https://github.com/openvinotoolkit/nncf/pull/1248.