get the named module in `mod` by the attribute name, for example ``look_up_named_module(net, "features.3.1.attn")`` Args: name:

Read image data from specified file or files, it can read a list of images and stack them together as multi-channel data in `get_data

Register ``ImageWriter``, so that writing a file with filename extension ``ext_name`` could be resolved to a tuple of potentially appropriate

Run a command by using ``subprocess.run`` with capture_output=True and stderr=subprocess.STDOUT so that the raise exception will have that in

From the mean and sigma representations resulting of encoding an image through the latent space, obtains a noise sample resulting fro

Save the state dict of input source data with PyTorch `save`. It can save `nn.Module`, `state_dict`, a dictionary of `nn.Module` or `state_di

Convert a string to a boolean. Case insensitive. True: yes, true, t, y, 1. False: no, false, f, n, 0. Args: value: string to be

Convert the input `data` endianness to `new`. Args: data: input to be converted. new: the target endianness, currently suppo

Append 1-sized dimensions to `arr` to create a result with `ndim` dimensions.

Run inference on `inputs` with the `network` model. Args: inputs: input of the model inference. network: mod

Args: include_background: if False, channel index 0 (background category) is excluded from the calculation. if th

Cosine similarity between L2-normalised class centroids. Args: emb: ``[N, D]`` float tensor. labels: ``[N]`` integer class labels

Linear CKA between source and target embedding matrices. Reference: Kornblith et al. (2019). Args: source: ``[N, D]`` float tensor.

This function is used to load pretrained models.

Add new analyzers to the engine so that the callable and summarize functions will utilize the new analyzers for stats computations.

Execute reduction and aggregation logic for the output of `compute_dice`. Args: reduction: defines mode of reduction as

Execute reduction logic for the output of `compute_panoptic_quality`. Args: reduction: define mode of reduction to the m

`np.allclose` with equivalent implementation for torch.

Temporarily set all MapTransforms to not throw an error if keys are missing. After, revert to original states. Args: transform: either Ma

Register a set of Ignite Event-Handlers to a specified Ignite engine. Args: engine: Ignite Engine, it can be a trainer,

Stop current running logger of MLFlow.

create a scaling matrix Args: spatial_dims: spatial rank scaling_factor: scaling factors for every spatial dim, defaults to

create a shearing matrix Args: spatial_dims: spatial rank coefs: shearing factors, a tuple of 2 floats for 2D, a tuple of 6

Error reporting method for CUDA calls. Args: cuda_ret: Tuple returned by CUDA runtime calls, where the first element is a

Fill the configuration templates, write the bundle (configs + scripts) to folder `output_path/algo_name`. Args: output_p

Generate the bundle scripts/configs for each bundleAlgo Args: output_folder: the output folder to save each algorithm.

Extract data array and metadata from loaded image and return them. This function returns two objects, first is numpy array of image d

Returns the algo output paths for scripts location

Create sinusoidal timestep embeddings following the implementation in Ho et al. "Denoising Diffusion Probabilistic Models" https://arxiv.org/

`torch.linalg.inv` with equivalent implementation for numpy. Args: x: array/tensor.

Apply pixel unshuffle to the tensor `x` with spatial dimensions `spatial_dims` and scaling factor `scale_factor`. Inverse operation of pixels

Evaluates the polynomial defined by `coef` at `x`. For a 1D sequence of coef (length n), evaluate:: y = coef[n-1] + x * (coef[n-2]

Pretty print the head and tail ``n_lines`` of ``val``, and omit the middle part if the part has more than 3 lines. Returns: the formatted st

Sometimes you need may to apply the same transform to different tensors. The idea is to get a sample and then apply it with apply() a

`np.ravel` with equivalent implementation for torch. Args: x: array/tensor to ravel. Returns: Return a contiguous flattened

MONAI seems to have different behavior in the borders of the image than ITK. This helper function sets the border of the ITK image as

Clear all the added `ConfigItem` and all the resolved content.

Automatically adjust the resampling interpolation mode and padding mode, so that they are compatible with the corresponding API of the `backe

Execute training based on Ignite Engine. If call this function multiple times, it will continuously run from the previous state.

Utility to safely convert the input data to target dtype. Args: data: input data can be PyTorch Tensor, numpy array, list, dictionar

Save the JIT object (script or trace produced object) `jit_obj` to the given file or stream with metadata included as a JSON file. The Torchs

Convert ``data_array`` into 'channel-last' numpy ndarray. Args: data_array: input data array with the channel dimension

Convert ``data_array`` into 'channel-last' numpy ndarray. Args: data_array: input data array with the channel dimension

Shut down the background thread for replacement.

Calculate gradients on single dimension of a tensor using central finite difference. It moves the tensor along the dimension to calculate the

Prepend 1-sized dimensions to `arr` to create a result with `ndim` dimensions.

Update cache items for current epoch, need to call this function before every epoch. If the cache has been shutdown before, need to r

Note that `torch.where` may convert y.dtype to x.dtype.

Generic function generator to replace base_t module with dest_t wrapper. Args: base_t : module type to replace dest_t : desti

Create a PIL image object from ``self.create_backend_obj(self.obj, ...)`` and call ``save``. Args: filename: filename or

Zero out the parameters of a module and return it.

Args: in_channels: dimension of input channels. out_channels: dimension of output channels. patch_size: s

Args: spatial_dims: number of spatial dimensions, could be 1, 2, or 3. in_channels: number of input channels.

Find the level corresponding to the value of the quantity in the list of values at each level. Args: name: the name of the request

This function computes the area (2D) or volume (3D) of each box. Half precision is not recommended for this function as it may cause overflow

Compute calibration bins for predicted probabilities and ground truth labels. This function implements hard binning for calibration analysis

Computes the Generalized Dice Score and returns a tensor with its per image values. Args: y_pred (torch.Tensor): Binarized segmentat

r""" This function computes the (Normalized) Surface Dice (NSD) between the two tensors `y_pred` (referred to as :math:`\hat{Y}`) and `y` (ref

Utility to convert a model into ONNX model and optionally verify with ONNX or onnxruntime. See also: https://pytorch.org/docs/stable/onnx.htm

Create an ITK object from ``data_array``. This method assumes a 'channel-last' ``data_array``. Args: data_array: input d

Specify `bundle workflow` to create monai bundle workflows. The workflow should be subclass of `BundleWorkflow` and be available to import.