Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[PT FE] Support aten::concatenate #28518

Merged
merged 1 commit into from
Jan 20, 2025
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions src/frontends/pytorch/src/op_table.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -425,6 +425,7 @@ const std::unordered_map<std::string, CreatorFunction> get_supported_ops_ts() {
{"aten::col2im", op::translate_col2im},
// aten::complex - Supported in limited set of patterns
{"aten::concat", op::translate_cat},
{"aten::concatenate", op::translate_cat},
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

how to reproduce such case in layer tests?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this is rare case, when something went wrong with normalization in tracing, I think this is difficult to catch such behaviour in layer test and we can trust to torch aliasing system that torch.concatenate is alias for torch.cat
https://github.com/pytorch/pytorch/blob/dbed747aae223d53ca4e22fe45c24d1d9a8b4432/torch/csrc/jit/passes/normalize_ops.cpp#L131

{"aten::contiguous", op::skip_node}, // In openvino how tensors are stored in memory is internal plugin detail,
// we assume all tensors are contiguous
{"aten::conv_transpose1d", op::translate_conv_transposend},
Expand Down
Loading