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.

Sign up for GitHub

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

Jump to bottom

[RFC][Discussion] Hidet Script #331

Closed
yaoyaoding opened this issue Jul 26, 2023 · 0 comments
Closed

[RFC][Discussion] Hidet Script #331

yaoyaoding opened this issue Jul 26, 2023 · 0 comments
Labels
documentation Improvements or additions to documentation rfc Discussion of potential rfc

Comments

@yaoyaoding
Copy link
Member

rendered draft.

Working in progress.
@yaoyaoding yaoyaoding added documentation Improvements or additions to documentation rfc Discussion of potential rfc labels Jul 30, 2023
vadiklyutiy pushed a commit that referenced this issue Jul 22, 2024
@BolinSNLHM @vadiklyutiy
[Fix] Fixing a bug in register_methods (#331)
7fbaa8d 
Previously, an error was encountered during a model compilation attempt:

> torch._dynamo.exc.BackendCompilerFailed: backend='hidet' raised:
> RuntimeError: Can not interpreting max given arguments:
>   max(tensor(...))
> Possible candidates are:
> torch_max_v3(x: hidet.Tensor, dim: Union[int, hidet.ir.expr.Expr],
keepdim: bool = False, *, out: Union[hidet.Tensor, Tuple[hidet.Tensor,
...], List[hidet.Tensor]] = None) -> Tuple[hidet.Tensor, hidet.Tensor]
> File
"/home/bolin/Desktop/hidet/python/hidet/graph/frontend/torch/register_functions.py",
line 1067

Despite we indeed have a
[function](https://github.com/CentML/hidet/blob/13a806608d40de2de1fcc682adeea8d204189f3c/python/hidet/graph/frontend/torch/register_functions.py#L1056-L1060)
that can be used to interpret the `torch.Tensor.max` with described
arguments.
vadiklyutiy pushed a commit that referenced this issue Jul 23, 2024
@BolinSNLHM @vadiklyutiy
[Fix] Fixing a bug in register_methods (#331)
c87c515 
Previously, an error was encountered during a model compilation attempt:

> torch._dynamo.exc.BackendCompilerFailed: backend='hidet' raised:
> RuntimeError: Can not interpreting max given arguments:
>   max(tensor(...))
> Possible candidates are:
> torch_max_v3(x: hidet.Tensor, dim: Union[int, hidet.ir.expr.Expr],
keepdim: bool = False, *, out: Union[hidet.Tensor, Tuple[hidet.Tensor,
...], List[hidet.Tensor]] = None) -> Tuple[hidet.Tensor, hidet.Tensor]
> File
"/home/bolin/Desktop/hidet/python/hidet/graph/frontend/torch/register_functions.py",
line 1067

Despite we indeed have a
[function](https://github.com/CentML/hidet/blob/13a806608d40de2de1fcc682adeea8d204189f3c/python/hidet/graph/frontend/torch/register_functions.py#L1056-L1060)
that can be used to interpret the `torch.Tensor.max` with described
arguments.
vadiklyutiy pushed a commit that referenced this issue Dec 26, 2024
@BolinSNLHM
[Fix] Fixing a bug in register_methods (#331)
6a2af72 
Previously, an error was encountered during a model compilation attempt:

> torch._dynamo.exc.BackendCompilerFailed: backend='hidet' raised:
> RuntimeError: Can not interpreting max given arguments:
>   max(tensor(...))
> Possible candidates are:
> torch_max_v3(x: hidet.Tensor, dim: Union[int, hidet.ir.expr.Expr],
keepdim: bool = False, *, out: Union[hidet.Tensor, Tuple[hidet.Tensor,
...], List[hidet.Tensor]] = None) -> Tuple[hidet.Tensor, hidet.Tensor]
> File
"/home/bolin/Desktop/hidet/python/hidet/graph/frontend/torch/register_functions.py",
line 1067

Despite we indeed have a
[function](https://github.com/CentML/hidet/blob/13a806608d40de2de1fcc682adeea8d204189f3c/python/hidet/graph/frontend/torch/register_functions.py#L1056-L1060)
that can be used to interpret the `torch.Tensor.max` with described
arguments.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
documentation Improvements or additions to documentation rfc Discussion of potential rfc
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@yaoyaoding