retinanet的focal loss注释版

主要写了对于focal loss如何具体地计算正负样本 及其iou loss的方式

class FocalLoss(nn.Module):
    #def __init__(self):

    def forward(self, classifications, regressions, anchors, annotations):
        alpha = 0.25
        gamma = 2.0
        batch_size = classifications.shape[0]
        classification_losses = []
        regression_losses = []
        # 对于不同batch 其anchor是一样的
        anchor = anchors[0, :, :]
        # anchors [N个目标,4] xyxy->xywh
        anchor_widths  = anchor[:, 2] - anchor[:, 0]
        anchor_heights = anchor[:, 3] - anchor[:, 1]
        anchor_ctr_x   = anchor[:, 0] + 0.5 * anchor_widths
        anchor_ctr_y   = anchor[:, 1] + 0.5 * anchor_heights

        for j in range(batch_size):
            # 提取每个batch的样本

            classification = classifications[j, :, :]
            regression = regressions[j, :, :]

            bbox_annotation = annotations[j, :, :]
            bbox_annotation = bbox_annotation[bbox_annotation[:, 4] != -1]
            #剔除异常值
            classification = torch.clamp(classification, 1e-4, 1.0 - 1e-4)
            #没有目标 纯背景的图片
            if bbox_annotation.shape[0] == 0:
                if torch.cuda.is_available():
                    alpha_factor = torch.ones(classification.shape).cuda() * alpha

                    alpha_factor = 1. - alpha_factor
                    focal_weight = classification
                    focal_weight = alpha_factor * torch.pow(focal_weight, gamma)

                    bce = -(torch.log(1.0 - classification))

                    # cls_loss = focal_weight * torch.pow(bce, gamma)
                    cls_loss = focal_weight * bce
                    classification_losses.append(cls_loss.sum())
                    regression_losses.append(torch.tensor(0).float().cuda())

                else:
                    alpha_factor = torch.ones(classification.shape) * alpha

                    alpha_factor = 1. - alpha_factor
                    focal_weight = classification
                    focal_weight = alpha_factor * torch.pow(focal_weight, gamma)

                    bce = -(torch.log(1.0 - classification))
                    # 仅计算分类损失 不计算回归损失
                    # cls_loss = focal_weight * torch.pow(bce, gamma)
                    cls_loss = focal_weight * bce
                    classification_losses.append(cls_loss.sum())
                    regression_losses.append(torch.tensor(0).float())

                continue
            #对每个anchor计算所有annotation的iou
            IoU = calc_iou(anchors[0, :, :], bbox_annotation[:, :4]) # num_anchors x num_annotations
            #计算每个anchor所对应的最大的iou
            IoU_max, IoU_argmax = torch.max(IoU, dim=1) # num_anchors x 1

            #import pdb
            #pdb.set_trace()

            # compute the loss for classification
            targets = torch.ones(classification.shape) * -1

            if torch.cuda.is_available():
                targets = targets.cuda()
            # iou小于0.4的判定为0
            targets[torch.lt(IoU_max, 0.4), :] = 0
            #iou大于0.5的判定为有目标 但未判定是否是正确的目标
            positive_indices = torch.ge(IoU_max, 0.5)

            num_positive_anchors = positive_indices.sum()
            #求出每个anchor 对应的gt
            assigned_annotations = bbox_annotation[IoU_argmax, :]
            # 对应的pos和种类置1
            targets[positive_indices, :] = 0
            targets[positive_indices, assigned_annotations[positive_indices, 4].long()] = 1

            if torch.cuda.is_available():
                alpha_factor = torch.ones(targets.shape).cuda() * alpha
            else:
                alpha_factor = torch.ones(targets.shape) * alpha

            alpha_factor = torch.where(torch.eq(targets, 1.), alpha_factor, 1. - alpha_factor)
            focal_weight = torch.where(torch.eq(targets, 1.), 1. - classification, classification)
            focal_weight = alpha_factor * torch.pow(focal_weight, gamma)
            # 将target与clssification结合 计算损失 相当于taregts是gt
            bce = -(targets * torch.log(classification) + (1.0 - targets) * torch.log(1.0 - classification))

            # cls_loss = focal_weight * torch.pow(bce, gamma)
            cls_loss = focal_weight * bce

            if torch.cuda.is_available():
                cls_loss = torch.where(torch.ne(targets, -1.0), cls_loss, torch.zeros(cls_loss.shape).cuda())
            else:
                cls_loss = torch.where(torch.ne(targets, -1.0), cls_loss, torch.zeros(cls_loss.shape))

            classification_losses.append(cls_loss.sum()/torch.clamp(num_positive_anchors.float(), min=1.0))

            # compute the loss for regression

            if positive_indices.sum() > 0:
                # 对于有iou>0.5的情况 也就是网络确信这边有目标的情况 计算reg loss

                assigned_annotations = assigned_annotations[positive_indices, :]

                anchor_widths_pi = anchor_widths[positive_indices]
                anchor_heights_pi = anchor_heights[positive_indices]
                anchor_ctr_x_pi = anchor_ctr_x[positive_indices]
                anchor_ctr_y_pi = anchor_ctr_y[positive_indices]

                gt_widths  = assigned_annotations[:, 2] - assigned_annotations[:, 0]
                gt_heights = assigned_annotations[:, 3] - assigned_annotations[:, 1]
                gt_ctr_x   = assigned_annotations[:, 0] + 0.5 * gt_widths
                gt_ctr_y   = assigned_annotations[:, 1] + 0.5 * gt_heights

                # clip widths to 1
                gt_widths  = torch.clamp(gt_widths, min=1)
                gt_heights = torch.clamp(gt_heights, min=1)
                # 求gt与真实值的误差大小 并作相关的归一化处理 归一化处理之后的值便是网络想要regression的值
                targets_dx = (gt_ctr_x - anchor_ctr_x_pi) / anchor_widths_pi
                targets_dy = (gt_ctr_y - anchor_ctr_y_pi) / anchor_heights_pi
                targets_dw = torch.log(gt_widths / anchor_widths_pi)
                targets_dh = torch.log(gt_heights / anchor_heights_pi)

                targets = torch.stack((targets_dx, targets_dy, targets_dw, targets_dh))
                targets = targets.t()

                if torch.cuda.is_available():
                    targets = targets/torch.Tensor([[0.1, 0.1, 0.2, 0.2]]).cuda()
                else:
                    targets = targets/torch.Tensor([[0.1, 0.1, 0.2, 0.2]])

                negative_indices = 1 + (~positive_indices)
                # 计算真实值与实际回归值的误差
                regression_diff = torch.abs(targets - regression[positive_indices, :])

                regression_loss = torch.where(
                    torch.le(regression_diff, 1.0 / 9.0),
                    0.5 * 9.0 * torch.pow(regression_diff, 2),
                    regression_diff - 0.5 / 9.0
                )
                regression_losses.append(regression_loss.mean())
            else:
                if torch.cuda.is_available():
                    regression_losses.append(torch.tensor(0).float().cuda())
                else:
                    regression_losses.append(torch.tensor(0).float())

        return torch.stack(classification_losses).mean(dim=0, keepdim=True), torch.stack(regression_losses).mean(dim=0, keepdim=True)