Preprocess an image for feature extraction. The length of the shorter edge is scaled to :obj:`self.min_size`. After the scaling, if the lengt

Use the given preset during prediction. This method changes values of :obj:`self.nms_thresh` and :obj:`self.score_thresh`. These valu

Get KDE estimators for two different datasets and a set of points to evaluate both distributions on.

Get KDE estimator for a given dataset, and generate a good set of points to sample the density at.

Get point estimates for KDE distributions for two different datasets.

Encodes the source and the destination bounding boxes to "loc". Given bounding boxes, this function computes offsets and scales to match the

create embedding dictionary for categorical features/columns

Returns the i-th example. Returns a color image and bounding boxes. The image is in CHW format. The returned image is RGB. A

Decode bounding boxes from bounding box offsets and scales. Given bounding box offsets and scales computed by :meth:`bbox2loc`, this function

Suppress bounding boxes according to their IoUs. This method checks each bounding box sequentially and selects the bounding box if the Inters

weight initalizer: truncated normal and random normal.

Detect objects from images. This method predicts objects for each image. Args: imgs (iterable of numpy.ndarray): Arrays

retrieve top k labels that occur most in all samples. Parameters ---------- data_dir: the directory that stores NUS-WIDE data.

wire up all models together as a single model for end-to-end training parameters: ---------- embedding_dict - the embedding dictiona