PyTorchCompressionEngine

CLASS - PyTorchCompressionEngine() :

Object that contains the Entry points to CLIKA Compression Engine.

Methods

deploy

deploy(

• clika_state_path: Union[str, Path]
• output_dir_path: Union[str, Path]
• file_suffix: Optional[str] =  None
• input_shapes: Union[list, tuple, List[list], List[tuple], Tuple[tuple], Tuple[list], None] =  None
• graph_author: Optional[str] =  None
• graph_description: Optional[str] =  None
• kwargs: Optional[dict] =  None

) ‑> str

Deploy the CLIKA State to the Format specified when the CLIKA state was created.

Parameters:

• clika_state_path (Union[str, Path]) - Path to the CLIKA state directory or file
• output_dir_path (Union[str, Path]) - Where to save the generated Model file
• file_suffix (Optional[str]) - Any string to prepend to the filename
• input_shapes (Optional[Union[list, tuple, List[list], List[tuple], Tuple[tuple], Tuple[list]]]) - Sequence or Sequence of Sequences that indicate the Input shape of the resulting Model. In-case of a single input model, you could pass tuple or a list such as: (1, 3, 224, 224) In-case of a multiple inputs model, you could pass [(1, 3, 224, 224), (1, 3, 120)] depending on your Model input shape that makes sense. 'None' value represents dynamic shape.
• graph_author (Optional[str]) - Optional string to specify who created the Graph for the Model on the resulting File
• graph_description (Optional[str]) - Optional string to specify Description for the Model on the resulting File
• kwargs (Optional[dict]) - Not used at the moment.

Returns:

• str: Path to the deployed Model

optimize

optimize(

• output_path: Union[str, Path]
• settings: Settings
• model: object
• model_compile_settings: ModelCompileSettings
• init_training_dataset_fn: Callable[[], Any]
• init_evaluation_dataset_fn: Optional[Callable[[], Any]]
• is_training_from_scratch: bool =  False
• multi_gpu: bool =  False
• kwargs: Optional[dict] =  None

) ‑> str

Entry point for running Compression.

Parameters:

• output_path (str/Path) - Path to save the files being generated by the library
• settings (Settings) - Settings that contains all the necessary information to run the model
• model (torch.nn.Module) - torch nn Module
• model_compile_settings (ModelCompileSettings) - The dataclass with the deep learning objects to use
• init_training_dataset_fn (Callable[[], Any]) - Pointer to a function that initializes the Dataloaders
• init_evaluation_dataset_fn (Optional[Callable[[], Any]]) - Pointer to a function that initializes the Dataloaders
• is_training_from_scratch (bool) - Whether or not the model is training from scratch
• multi_gpu (bool) - Whether or not to use Multi-GPU.
• kwargs (Optional[dict]) - Not used at the moment

Returns:

• str: Path to the latest CLIKA state

resume

resume(

• clika_state_path: Union[str, Path]
• model_compile_settings: ModelCompileSettings
• init_training_dataset_fn: Callable[[], Any]
• init_evaluation_dataset_fn: Optional[Callable[[], Any]]
• settings: Settings =  None
• multi_gpu: bool =  False
• resume_epoch: Optional[int] =  None
• kwargs: Optional[dict] =  None

) ‑> str

Resumes a Compression Job from Given path. This can be used to finetune on more Data or just resume a run that had to be stopped.

Parameters:

• clika_state_path (Union[str, Path]) - Path to the CLIKA state directory or file
• model_compile_settings (ModelCompileSettings) - The dataclass with the deep learning objects to use
• init_training_dataset_fn (Callable[[], Any]) - Pointer to a function that initializes the Dataloaders
• init_evaluation_dataset_fn (Optional[Callable[[], Any]]) - Pointer to a function that initializes the Dataloaders
• settings (Settings) - object to use for the Compression Run. TrainingSettings and DeploymentSettings will be updated with this.
• multi_gpu (bool) - whether or not to do Multi-GPU compression
• resume_epoch (Optional[int]) - Will override the Epoch present in the CLIKA State directory at Path.
• kwargs (Optional[dict]) - Unsupported for now.

Returns:

• str: Path to latest CLIKA State path