direct.nn.jointicnet package#

Submodules#

direct.nn.jointicnet.config module#

class direct.nn.jointicnet.config.JointICNetConfig(model_name: str = '???', engine_name: str | None = None, num_iter: int = 10, use_norm_unet: bool = False, image_unet_num_filters: int = 8, image_unet_num_pool_layers: int = 4, image_unet_dropout: float = 0.0, kspace_unet_num_filters: int = 8, kspace_unet_num_pool_layers: int = 4, kspace_unet_dropout: float = 0.0, sens_unet_num_filters: int = 8, sens_unet_num_pool_layers: int = 4, sens_unet_dropout: float = 0.0)[source][source]#

Bases: ModelConfig

image_unet_dropout: float = 0.0#
image_unet_num_filters: int = 8#
image_unet_num_pool_layers: int = 4#
kspace_unet_dropout: float = 0.0#
kspace_unet_num_filters: int = 8#
kspace_unet_num_pool_layers: int = 4#
num_iter: int = 10#
sens_unet_dropout: float = 0.0#
sens_unet_num_filters: int = 8#
sens_unet_num_pool_layers: int = 4#
use_norm_unet: bool = False#

direct.nn.jointicnet.jointicnet module#

class direct.nn.jointicnet.jointicnet.JointICNet(forward_operator, backward_operator, num_iter=10, use_norm_unet=False, **kwargs)[source][source]#

Bases: Module

Joint Deep Model-Based MR Image and Coil Sensitivity Reconstruction Network (Joint-ICNet) implementation as presented in [1].

References

[1]

Jun, Yohan, et al. “Joint Deep Model-Based MR Image and Coil Sensitivity Reconstruction Network (Joint-ICNet) for Fast MRI.” 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2021, pp. 5266–75. DOI.org (Crossref), https://doi.org/10.1109/CVPR46437.2021.00523.

forward(masked_kspace, sampling_mask, sensitivity_map)[source][source]#

Computes forward pass of JointICNet.

Parameters:
masked_kspace: torch.Tensor

Masked k-space of shape (N, coil, height, width, complex=2).

sampling_mask: torch.Tensor

Sampling mask of shape (N, 1, height, width, 1).

sensitivity_map: torch.Tensor

Sensitivity map of shape (N, coil, height, width, complex=2).

Returns:
out_image: torch.Tensor

Output image of shape (N, height, width, complex=2).

Return type:

Tensor

training: bool#

direct.nn.jointicnet.jointicnet_engine module#

class direct.nn.jointicnet.jointicnet_engine.JointICNetEngine(cfg, model, device, forward_operator=None, backward_operator=None, mixed_precision=False, **models)[source][source]#

Bases: MRIModelEngine

Joint-ICNet Engine.

forward_function(data)[source][source]#

This method performs the model’s forward method given data which contains all tensor inputs.

Must be implemented by child classes.

Return type:

Tuple[Tensor, None]

Module contents#