tenstorrent / tt-mlir

Tenstorrent MLIR compiler
https://tenstorrent.github.io/tt-mlir/
Apache License 2.0
59 stars 7 forks source link

Add squeeze op from 4D to input rank for Transpose #729

Closed dgolubovicTT closed 1 week ago

dgolubovicTT commented 1 week ago

Add squeeze op from 4D to input rank for Transpose, because we used unsqueeze_to_4D and we shouldn't leave tensor with changed rank after Transpose op.

dgolubovicTT commented 1 week ago

solved by https://github.com/tenstorrent/tt-mlir/pull/730