Save model checkpoint. :param epoch: epoch number :param model: model :param optimizer: optimizer

Run one train epoch

One epoch's training. :param train_loader: DataLoader for training data :param model: model :param criterion: MultiBox loss :par

Apply the transformations above. :param image: image, a PIL Image :param boxes: bounding boxes in boundary coordinates, a tensor of dime

Convert bounding boxes from boundary coordinates (x_min, y_min, x_max, y_max) to center-size coordinates (c_x, c_y, w, h). :param xy: boundi

norm: loss的归一化系数，用batch中所有有效token数即可

norm: loss的归一化系数，用batch中所有有效token数即可

norm: loss的归一化系数，用batch中所有有效token数即可

获取对应index的图像和ground truth label，并视情况进行数据增强

获取对应index的图像和ground truth label，并视情况进行数据增强

Take in model size and number of heads.

Take in model size and number of heads.

Take in model size and number of heads.

位置编码器类的初始化函数 共有三个参数，分别是 d_model：词嵌入维度 dropout: dropout触发比率 max_len：每个句子的最大长度

:param data_folder: folder where data files are stored :param split: split, one of 'TRAIN' or 'TEST' :param keep_difficult: k

:param n_classes: number of different types of objects

Computes top-k accuracy, from predicted and true labels. :param scores: scores from the model :param targets: true labels :param k:

Since each image may have a different number of objects, we need a collate function (to be passed to the DataLoader). This describes

Take in and process masked src and target sequences.