torch_api Module Reference

Functions/Subroutines
subroutine, public	torch_dict_create (dict)
	Creates an empty Torch dictionary. More...
subroutine, public	torch_dict_release (dict)
	Releases a Torch dictionary and all its ressources. More...
subroutine, public	torch_model_load (model, filename)
	Loads a Torch model from given "*.pth" file. (In Torch lingo models are called modules) More...
subroutine, public	torch_model_eval (model, inputs, outputs)
	Evaluates the given Torch model. (In Torch lingo this operation is called forward()) More...
subroutine, public	torch_model_release (model)
	Releases a Torch model and all its ressources. More...
character(:) function, allocatable, public	torch_model_read_metadata (filename, key)
	Reads metadata entry from given "*.pth" file. (In Torch lingo they are called extra files) More...
logical function, public	torch_cuda_is_available ()
	Returns true iff the Torch CUDA backend is available. More...
subroutine, public	torch_allow_tf32 (allow_tf32)
	Set whether to allow the use of TF32. Needed due to changes in defaults from pytorch 1.7 to 1.11 to >=1.12 See https://pytorch.org/docs/stable/notes/cuda.html. More...
subroutine, public	torch_model_freeze (model)
	Freeze the given Torch model: applies generic optimization that speed up model. See https://pytorch.org/docs/stable/generated/torch.jit.freeze.html. More...

Function/Subroutine Documentation

torch_dict_create()

subroutine, public torch_api::torch_dict_create

(

type(torch_dict_type), intent(inout)

dict

)

Creates an empty Torch dictionary.

Author

Ole Schuett

Definition at line 895 of file torch_api.F.

Here is the caller graph for this function:

torch_dict_release()

subroutine, public torch_api::torch_dict_release

(

type(torch_dict_type), intent(inout)

dict

)

Releases a Torch dictionary and all its ressources.

Author

Ole Schuett

Definition at line 919 of file torch_api.F.

Here is the caller graph for this function:

torch_model_load()

subroutine, public torch_api::torch_model_load	(	type(torch_model_type), intent(inout)	model,
		character(len=*), intent(in)	filename
	)

Loads a Torch model from given "*.pth" file. (In Torch lingo models are called modules)

Author

Ole Schuett

Definition at line 943 of file torch_api.F.

Here is the caller graph for this function:

torch_model_eval()

subroutine, public torch_api::torch_model_eval	(	type(torch_model_type), intent(inout)	model,
		type(torch_dict_type), intent(in)	inputs,
		type(torch_dict_type), intent(inout)	outputs
	)

Evaluates the given Torch model. (In Torch lingo this operation is called forward())

Author

Ole Schuett

Definition at line 970 of file torch_api.F.

Here is the caller graph for this function:

torch_model_release()

subroutine, public torch_api::torch_model_release

(

type(torch_model_type), intent(inout)

model

)

Releases a Torch model and all its ressources.

Author

Ole Schuett

Definition at line 1003 of file torch_api.F.

Here is the caller graph for this function:

torch_model_read_metadata()

character(:) function, allocatable, public torch_api::torch_model_read_metadata	(	character(len=*), intent(in)	filename,
		character(len=*), intent(in)	key
	)

Reads metadata entry from given "*.pth" file. (In Torch lingo they are called extra files)

Author

Ole Schuett

Definition at line 1027 of file torch_api.F.

Here is the caller graph for this function:

torch_cuda_is_available()

logical function, public torch_api::torch_cuda_is_available

Returns true iff the Torch CUDA backend is available.

Author

Ole Schuett

Definition at line 1079 of file torch_api.F.

Here is the caller graph for this function:

torch_allow_tf32()

subroutine, public torch_api::torch_allow_tf32

(

logical, intent(in)

allow_tf32

)

Set whether to allow the use of TF32. Needed due to changes in defaults from pytorch 1.7 to 1.11 to >=1.12 See https://pytorch.org/docs/stable/notes/cuda.html.

Author

Gabriele Tocci

Definition at line 1103 of file torch_api.F.

Here is the caller graph for this function:

torch_model_freeze()

subroutine, public torch_api::torch_model_freeze

(

type(torch_model_type), intent(inout)

model

)

Freeze the given Torch model: applies generic optimization that speed up model. See https://pytorch.org/docs/stable/generated/torch.jit.freeze.html.

Author

Gabriele Tocci

Definition at line 1126 of file torch_api.F.

Here is the caller graph for this function: