Skip to content

[Device] Add support for mixed cpu and cuda kernels in the same flow graph#270

Merged
yaoyaoding merged 5 commits intohidet-org:mainfrom
yaoyaoding:transfer-ops
Jun 5, 2023
Merged

[Device] Add support for mixed cpu and cuda kernels in the same flow graph#270
yaoyaoding merged 5 commits intohidet-org:mainfrom
yaoyaoding:transfer-ops

Conversation

@yaoyaoding
Copy link
Copy Markdown
Member

No description provided.

@yaoyaoding yaoyaoding merged commit 9cbbf6d into hidet-org:main Jun 5, 2023
@yaoyaoding yaoyaoding deleted the transfer-ops branch June 5, 2023 02:56
vadiklyutiy pushed a commit that referenced this pull request Jul 22, 2024
Closes #222 

Also made some small changes to the function `torch_pad` in
`register_functions.py` to make it have the same behavior as
`torch.nn.functional.pad` when the keyword argument `value` is its
default value `None`.

In the interest of time, I did not implement
`torch.Tensor.repeat_interleave` with the argument `repeats` being a
tensor, because it is not encountered yet.


[Edit: Also added support for `torch.Tensor.unbind`, which was
encountered while running the same model, immediately after adding
support for previous ones. ]
vadiklyutiy pushed a commit that referenced this pull request Jul 23, 2024
Closes #222 

Also made some small changes to the function `torch_pad` in
`register_functions.py` to make it have the same behavior as
`torch.nn.functional.pad` when the keyword argument `value` is its
default value `None`.

In the interest of time, I did not implement
`torch.Tensor.repeat_interleave` with the argument `repeats` being a
tensor, because it is not encountered yet.


[Edit: Also added support for `torch.Tensor.unbind`, which was
encountered while running the same model, immediately after adding
support for previous ones. ]
vadiklyutiy pushed a commit that referenced this pull request Dec 26, 2024
Closes #222 

Also made some small changes to the function `torch_pad` in
`register_functions.py` to make it have the same behavior as
`torch.nn.functional.pad` when the keyword argument `value` is its
default value `None`.

In the interest of time, I did not implement
`torch.Tensor.repeat_interleave` with the argument `repeats` being a
tensor, because it is not encountered yet.


[Edit: Also added support for `torch.Tensor.unbind`, which was
encountered while running the same model, immediately after adding
support for previous ones. ]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant