NRTRModalityTransform

class mmocr.models.textrecog.NRTRModalityTransform(in_channels=3, init_cfg=[{'type': 'Kaiming', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}])

Modality transform in NRTR.

参数

• in_channels (int) – Input channel of image. Defaults to 3.

• init_cfg (dict or list[dict], optional) – Initialization configs.

返回类型

None

forward(x)

Backbone forward.

参数

x (torch.Tensor) – Image tensor of shape \((N, C, W, H)\). W, H is the width and height of image.

返回

Output tensor.

返回类型

Tensor