Returns the noise tensor for input tensor `images`, override this for different noise distributions.

Get number of possible devices detected by OpenCV that can be used for capture.

Sample enough negatives to fill up ``self.batch_size_per_image`` Args: negative: indices of positive samples

Number of positive samples to draw Args: positive: indices of positive samples Returns: number of p

Read the optional package versions into a dictionary.

Return the actual spatial shape covered by the output split patches. For instance, if the input image is padded, the actual spatial s

Returns the objective power of the whole slide image at a given level. Args: wsi: a whole slide image object loaded from

Get the loaded properties of dataset with specified keys. If no keys specified, return all the loaded properties.

Get the seed and label arrays for the max-flow algorithm Args: data: Input array sink_mode (str, optional): Sink mode

Returns the SEBlock defined in the initialization of the QuickNAT model. Args: n_filters: encoding half of the layer

Returns the size (height, width) of the whole slide image at a given level. Args: wsi: a whole slide image object loaded

Get the 3 orthogonal views and stack them into 1 image. Requires that all images be same size, but this is taken care of by the `SpatialPadd`

Get system info as an ordered dictionary.

Return the target for the loss function, this is the `noise` value by default.

Gets the ground truth tensor for the discriminator depending on whether the input is real or fake. Args: input: input te

Get the identifier of the current experiment. In the format of listing the searching parameter name and values connected by underscor

Get the identifier of the current experiment. In the format of listing the searching parameter name and values connected by underscor

Returns the same information as `get_times_summary` but in a Pandas DataFrame.

Get a timestep, by default this is a random integer between 0 and `self.num_train_timesteps`.

Get topology entropy loss at searching stage. Args: probs: path activation probabilities

Returns the level associated to the resolution parameters in the whole slide image. Args: wsi: a whole slide image objec

Get the accumulated values during aggregation or final averaged values after it is finalized. Returns: torch.tensor: agg

Get the accumulated values during aggregation Returns: zarr.Array: aggregated values.

Calculates weights for a given number of scale levels

Returns the representation of the whole slide image at a given micro-per-pixel (mpp) resolution. The optional tolerance parameters ar

Get version from 'git describe' in the root of the source tree. This only gets called if the git-archive 'subst' keywords were *not* expanded

Get version information from git keywords.

Check if there is any Ampere and later GPU.

Returns True if the user has installed a version of cupy.

whether the current environment has InstanceNorm3dNVFuser https://github.com/NVIDIA/apex/blob/23.05-devel/apex/normalization/instance_norm.py#L15-

ICNR initialization for 2D/3D kernels adapted from Aitken et al.,2017 , "Checkerboard artifact free sub-pixel convolution".

Import cv2. Put it inside a function to avoid webcam lights blinking on ``import monai``.

`np.in1d` with equivalent implementation for torch.

Initialize routine to parse configuration files and extract main components such as trainer, evaluator, and filters. Args:

Initialize the bundle workflow before running.

Returns number of folds in the datalist file, and assigns fold numbers if not provided. Args: datalist_filename: path to

Create an object instance or call a callable object from a class or function represented by ``_path``. `kwargs` will be part of the input arg

Check if two sets of kwargs are compatible (to be combined in `apply`).

Utility function used in `instantiate()` to check whether to skip the instantiation.

Determine if the object is an immutable object. see also https://github.com/python/cpython/blob/3.11/Lib/copy.py#L109

Determine if the input is a square number or a squence of square numbers.

Returns True if `name` is a valid Python variable name and also not a keyword.

`np.isfinite` with equivalent implementation for torch.

Yield successive tuples of slices defining patches of size `patch_size` from an array of dimensions `image_size`. The iteration starts from p

Iterate over chunks of a given shape. Args: chunks: the chunk shape cdata_shape: the shape of the data in chunks sli

warping with python itk Args: img: numpy array of shape (D, H, W) ddf: numpy array of shape (3, D, H, W) Returns:

Keep connected regions that include the positive points. Used for point-only inference postprocessing to remove regions without positive poin

Compute the union of class IDs in label and generate a list to include all class IDs Args: x: a list of numbers (for example, class_I

Learns, from scratch, the distribution of each class from the provided labels. Args: features (torch.Tensor): features f

To convert a list of "key=value" pairs into a dictionary. For examples: items: `["a=1", "b=2", "c=3"]`, return: {"a": "1", "b": "2", "c": "3"

Convert a list of arguments to a string that can be used in python-fire. Args: args: the list of arguments. Returns: th

Extract the properties with the specified keys from the Decathlon JSON file. See under `load_decathlon_datalist` for the expected keys in the Deca

Download and extract Medical Model Archive (MMAR) model weights from Nvidia Clara Train. Args: item: the corresponding model item fr

Load a state dict from a DiffusionModelUNet trained with [MONAI Generative](https://github.com/Project-MONAI/GenerativeModels).

Load a state dict from a ControlNet trained with [MONAI Generative](https://github.com/Project-MONAI/GenerativeModels). Args

Load a state dict from a ViTAutoEnc model trained with an older version of MONAI where ``CrossAttentionBlock`` was unconditionally in

Load a state dict from a DecoderOnlyTransformer trained with [MONAI Generative](https://github.com/Project-MONAI/GenerativeModels).

Load a state dict from a MaskedAutoEncoderViT model trained with an older version of MONAI where ``CrossAttentionBlock`` was uncondit

Log metrics from the given Engine's state member. Args: engine: Ignite Engine to log from

Combine "map_classes_to_indices" and "generate_label_classes_crop_centers" functions, return crop center coordinates. This calls `map_classes

Utility to map the spatial axes to real axes in channel first/last shape. For example: If `channel_first` is True, and `img` has 3 spatia

Match the SeriesInstanceUID from all series in a study according to the input SOPInstanceUID. Args: study_uid: StudyInstanceUID.

Match boxes of a batch to corresponding ground truth for each category independently. Args: iou_fn: compute overlap for each pai

`np.maximum` with equivalent implementation for torch. Args: a: first array/tensor. b: second array/tensor. Returns: