Closed keerthana-r-mcw closed 3 months ago
@keerthana-r-mcw and @saichandax please try to find out which op is causing the error? Include the unit test for the specific op that is generating this error in this issue.
Hi,
The error is occuring in Conv2d(576, 160, 1, 1, bias=False)
at the line
models/experimental/functional_mobilenetv2/tt/ttnn_mobilenetv2.py:277: in __call__ output_tensor_c42 = self.c42(output_tensor)
.
Also pushed the unit test for that case.
Issue is resolved on the latest main.
Mobilenetv2 pipeline fails with the error "Grid is invalid for mcast matmul!"
To Reproduce Steps to reproduce the behavior:
pytest tests/ttnn/integration_tests/mobilenetv2/test_ttnn_mobilenetv2.py
RuntimeError: TT_FATAL @ tt_eager/tt_dnn/op_library/bmm/multi_core_reuse_mcast_2d_optimized/bmm_op_multi_core_reuse_mcast_2d_optimized.cpp:1037: false E info: E Grid is invalid for mcast matmul!
pytest test/ttnn/unit_tests/operations/test_conv2d.py:: test_mobilenetv2_conv_ttnn_grid_issue_case