Replaces deprecated (pre python 3.12) distutils strtobool function. True values are y, yes, t, true, on and 1; False values are n, n

This function swaps nested modules as specified by "dot paths" in mod with a desired replacement. This allows for swapping nested modules thr

For compatibility with nibabel, switch from LPS to RAS. Adapt affine matrix and `space` argument in header accordingly. If no informa

Implementation based on native PyTorch distributed data parallel APIs.

Train the Algos using HPO. Args: history: the history of generated Algos. It is a list of dicts. Each element has the ta

Train the Algos in a sequential scheme. The order of training is randomized. Args: history: the history of generated Alg

Transform the randomly selected `self.hole_coords` in input images.

Wait thread lock and replace training items in the background thread.

Wait for thread lock to shut down the background thread.

Update the cache items with new replacement for current epoch.

Set spatial_shape explicitly from resolved shape.

Abort the training or evaluation. Args: extra: Dict with additional information that can be provided by the FL system.

Creates an animated gif out of an image tensor in 'CHWD' format and writes it with SummaryWriter. Args: writer: Tensorboard SummaryWriter

Top-level function to add cast wrappers around modules known to cause issues for FP16/autocast ONNX export NOTE: This occurs in place, if you

r""" Calculate decomposed Relative Positional Embeddings from mvitv2 implementation: https://github.com/facebookresearch/mvit/blob/19786631e33

Returns a decorator which stores the decorated function under `name` with description `desc`.

Add model inferer to the builder. Args: identifier: name of the bundleAlgo. gen_algo: a trained BundleAlgo m

Add noise to the original samples. Args: original_samples: original samples noise: noise to add to samples

Add a result in a thread-safe manner to the internal results dictionary.

Get the affine. Defaults to ``torch.eye(4, dtype=torch.float64)``

Import the Algo object from a pickle file. **Unsafe**; prefer ``algo_from_json``. Disabled unless ``MONAI_ALLOW_PICKLE=1`` is set. See ``_load_le

Export the Algo object to a pickle file. **Unsafe**; prefer ``algo_to_json``. Pickle can execute arbitrary code on load. This function is disable

Allocates outputs to run TRT engine Args: device: GPU device to allocate memory on

If true, Auto3DSeg algo (de)serialization may use pickle. Default False. Pickle can execute arbitrary code on load and should only be enabled

Analyze (training) data Args: datalist_json: original data list .json (required by most monai tutorials). data_dir: raw data

apply affine transformation to a set of points. Args: data: input data to apply affine transformation, should be a tensor of shape (

This method causes "in order" processing of pending transforms to occur. "in order" processing of pending transforms ensures that all pending

`np.argwhere` with equivalent implementation for torch. Args: a: input data. Returns: Indices of elements that are non-zero.

Args: engine: Ignite Engine, it can be a trainer, validator or evaluator.

Args: engine: Ignite Engine, it can be a trainer, validator or evaluator.

Args: engine: Ignite Engine, it can be a trainer, validator or evaluator.

Attach event handlers to the given evaluator to log metric values from it. Args: evaluator: Ignite Engine implementing

Compute attenuation weighting Args: img (NDArray): Image alpha: Attenuation coefficient (see publication) Re

Performs non-maximum suppression in a batched fashion. Each labels value correspond to a category, and NMS will not be applied between elemen

Compute Binary CrossEntropy loss for the input logits and target in one single class.

Blend an image and a label. Both should have the shape CHW[D]. The image may have C==1 or 3 channels (greyscale or RGB). The label is exp

Compute the intersection over union (IoU) of two set of boxes. Args: boxes1: bounding boxes, Nx4 or Nx6 torch tensor or ndarray. The

Create a ThreadBuffer object using the `src`, `buffer_size`, and `timeout` parameters given for the constructor arguments of the same names,

Builds a (Anistropic) Fourier feature position embedding based on the given grid size, embed dimension, spatial dimensions, and scales. The s

This function is used to calculate the percentiles of intensities (and median) of the input dataset. To get the required values, all

This function is used to calculate the maximum, minimum, mean and standard deviation of intensities of the input dataset. Ar

Utility function to cast a single tensor from from_dtype to to_dtype

Check the operations of a MetaTensor to determine whether there are any statuses Args: entry: a dictionary that may contain TraceKey.

There are many metrics related to confusion matrix, and some of the metrics have more than one names. In addition, some of the names are very

We expect the values in ``head_outputs``: Dict[str, List[Tensor]] to have the same length. Will raise ValueError if not. Args: h

Validate the input dimensionality (raise a `ValueError` if invalid). Args: input_images: It can be 1) a tensor sized (B, C, H, W) or

Check if shape of input in first dimension is consistent with expected IoU values (assumes IoU dimension is the first dimension)

Validate the input images/targets during training (raise a `ValueError` if invalid). Args: input_images: It can be 1) a tensor sized

Check whether a traced transform entry matches this transform. When multiprocessing uses ``spawn``, transform instances are recreated,

Collate a sequence of meta tensor into a single batched metatensor. This is called by `collage_meta_tensor` and so should not be used as a co

This function is used to collect the metadata for all images of the dataset.

Given transforms A and B to be applied to x, return the combined transform (AB), so that A(B(x)) becomes AB(x)