r"""Learning Rate Scheduler Parameters ---------- mode : str Modes for learning rate scheduler. Currently it supports 'c

Pre-trained ResNetV1b Model, which produces the strides of 8 featuremaps at conv5. Parameters ---------- block : Block Class

r"""Stack the input data samples to construct the batch. The N input samples must have the same shape/length and will be stacked to construct a ba

Wrap multiple batchify functions to form a function apply each input function on each input fields respectively. Each data sample should be a

r"""ResNet V2 model from `"Identity Mappings in Deep Residual Networks" <https://arxiv.org/abs/1603.05027>`_ paper. Parameters ------

Pad the input ndarrays along the specific padding axis and stack them to get the output. Input of the function will be N samples. Each sample shou

r"""Loosely return list of the input data samples. There is no constraint of shape for any of the input samples, however, you will only be abl

Feature extractor with additional layers to append. This is specified for ``Feature Pyramid Network for Object Detection`` which implement ``T

Convert corner boxes to center boxes. Corner boxes are encoded as (xmin, ymin, xmax, ymax) Center boxes are encoded as (center_x, center_y, wi

Detection metric for COCO bbox task. Parameters ---------- dataset : instance of gluoncv.data.COCODetection The validation datase

r""" Parameters ---------- size : int Upsampling size.

Mean average precision metric for PASCAL V0C 07 dataset Parameters: --------- iou_thresh : float IOU overlap threshold for TP

Pascal VOC detection Dataset. Parameters ---------- root : str, default '~/mxnet/datasets/voc' Path to folder storing the dataset

Data parallelism Hide the difference of single/multiple GPUs to the user. Inputs and outputs are both list of NDArrays in different contexts.

r"""Compose Learning Rate Schedulers Parameters ---------- schedulers: list list of LRScheduler objects

SoftmaxCrossEntropyLoss2D with Auxiliary Loss Parameters ---------- aux : bool, default True Whether to use auxiliary loss. a

Criterion data parallelism Parameters ---------- module : object Network to be parallelized. ctx : list A list of con

Load the Market 1501 dataset. Parameters ---------- items : list List for image names and labels. flag : int, default 1

Decode classification results. This decoder must work with `MultiClassEncoder` to reconstruct valid labels. The decoder expect results are af

ADE20K Semantic Segmentation Dataset. Parameters ---------- root : string Path to VOCdevkit folder. Default is '$(HOME)/mxnet/dat

Convert center boxes to corner boxes. Corner boxes are encoded as (xmin, ymin, xmax, ymax) Center boxes are encoded as (center_x, center_y, wi

Split bounding boxes into 4 columns. Parameters ---------- axis : int, default is -1 On which axis to split the bounding box. Def

Convolutional predictor. Convolutional predictor is widely used in object-detection. It can be used to predict classification scores (1 channe

Denormalize the image

Decode bounding boxes training target with normalized center offsets. This decoder must cooperate with NormalizedBoxCenterEncoder of same `stds`

RPN target generator network. Parameters ---------- num_sample : int, default is 256 Number of samples for RPN targets. pos_i

DataLoader that support random transform function applied to dataset. Parameters ---------- transform_fns : iterable of callables

Resize an image to the given size. Parameters ---------- size : int or tuple of (W, H) Size of output image. keep_ratio : boo

r"""Training History Record and Plot Parameters ---------- labels : list of str List of names of the labels in the history.

Clip bounding box to image edges. Parameters ---------- axis : int The coordinate axis with length 4.

r"""Bottleneck Block from `"Aggregated Residual Transformations for Deep Neural Network" <http://arxiv.org/abs/1611.05431>`_ paper. Parameter

r"""Bottleneck Block from `"Aggregated Residual Transformations for Deep Neural Networks" <http://arxiv.org/abs/1611.05431>`_ paper. Paramete

MS COCO instance segmentation dataset. Parameters ---------- root : str, default '~/mxnet/datasets/coco' Path to folder storing t

Instance segmentation metric for COCO bbox and segm task. Will return box summary, box metric, seg summary and seg metric. Parameters ---

Cityscapes Dataloader

Instance segmentation metric for COCO bbox and segm task. Will return box summary, box metric, seg summary and seg metric. Parameters ---

Feature extractor with additional layers to append. This is very common in vision networks where extra branches are attached to backbone netwo

If the batch size is small, it's better to use GroupNorm instead of BatchNorm. GroupNorm achieves good results even at small batch sizes.

Computes accuracy classification score with optional ignored labels. The accuracy score is defined as .. math:: \\text{accuracy}(y, \\

Load the ImageNet classification dataset. Refer to :doc:`../build/examples_datasets/imagenet` for the description of this dataset and how to

Detection dataset loaded from LST file and raw images. LST file is a pure text file but with special label format. Checkout :ref:`lst_record_

Mask RCNN target encoder to generate mask targets. Parameters ---------- num_images : int Number of input images. num_rois :

Encode classification training target given matching results. This encoder will assign training target of matched bounding boxes to ground-tr

Encode bounding boxes training target with normalized center offsets. Input bounding boxes are using corner type: `x_{min}, y_{min}, x_{max}, y_{

RCNN target encoder to generate matching target and regression target values. Parameters ---------- num_class : int Number of tot

r"""Anchor generator for Region Proposal Networks. Parameters ---------- stride : int Feature map stride with respect to original

Resize an image to the given size. Parameters ---------- size : int or tuple of (W, H) Size of output image. keep_ratio : boo

r"""SEBlock from `"Aggregated Residual Transformations for Deep Neural Network" <http://arxiv.org/abs/1611.05431>`_ paper. Parameters ---

Default SSD training transform which includes tons of image augmentations. Parameters ---------- width : int Image width. hei

Default SSD validation transform. Parameters ---------- width : int Image width. height : int Image height. mean

Training targets generator for Single-shot Object Detection. Parameters ---------- iou_thresh : float IOU overlap threshold for m

Segmentation Eval Module

Default training transform for simple pose. Parameters ---------- num_joints : int Number of joints defined by dataset image_

Default YOLO training transform which includes tons of image augmentations. Parameters ---------- width : int Image width. he

Default YOLO validation transform. Parameters ---------- width : int Image width. height : int Image height. mean

YOLOV3 target merger that merges the prefetched targets and dynamic targets. Parameters ---------- num_class : int Number of fore

/ * * @brief predicate that returns true if the input string does not start * with a given prefix * ***************************************

r"""AlexNet model from the `"One weird trick..." <https://arxiv.org/abs/1404.5997>`_ paper. Parameters ---------- classes : int, default

Batch Bounding Box IOU. Parameters ---------- axis : int On which axis is the length-4 bounding box dimension. fmt : str

A Matcher implementing bipartite matching strategy. Parameters ---------- threshold : float Threshold used to ignore invalid padd

r"""ResNext model from `"Aggregated Residual Transformations for Deep Neural Networks" <http://arxiv.org/abs/1611.05431>`_ paper. Parameters

r"""ResNet V2 model from `"Identity Mappings in Deep Residual Networks" <https://arxiv.org/abs/1603.05027>`_ paper. Parameters ------

Detection metric for COCO bbox task. Parameters ---------- dataset : instance of gluoncv.data.COCODetection The validation datase

A Matcher that combines multiple strategies. Parameters ---------- matchers : list of Matcher Matcher is a Block/HybridBlock used

Darknet Basic Block. Which is a 1x1 reduce conv followed by 3x3 conv. Parameters ---------- channel : int Convolution channels fo

r"""DeepLabV3 Parameters ---------- nclass : int Number of categories for the training dataset. backbone : string Pre

r"""Densenet-BC model from the `"Densely Connected Convolutional Networks" <https://arxiv.org/pdf/1608.06993.pdf>`_ paper. Parameters ---

Data loader for detection dataset. .. deprecated:: 0.2.0 :py:class:`DetectionDataLoader` is deprecated, please use :py:class:`mxn