mipcandy.data.convertion

Module Contents

Functions

convert_ids_to_logits	param ids: class ids (…, 1, …) param num_classes: number of classes param channel_dim: the index of the channel dimension return: logits (…, num_classes, …)
convert_logits_to_ids	param logits: logits (…, num_classes, …) param channel_dim: the index of the channel dimension return: class ids (…, 1, …)
auto_convert

API

mipcandy.data.convertion.convert_ids_to_logits(ids: torch.Tensor, num_classes: int, *, channel_dim: int = 1) → torch.Tensor

Parameters:

• ids – class ids (…, 1, …)

• num_classes – number of classes

• channel_dim – the index of the channel dimension

Returns:

logits (…, num_classes, …)

mipcandy.data.convertion.convert_logits_to_ids(logits: torch.Tensor, *, channel_dim: int = 1) → torch.Tensor

Parameters:

• logits – logits (…, num_classes, …)

• channel_dim – the index of the channel dimension

Returns:

class ids (…, 1, …)

mipcandy.data.convertion.auto_convert(image: torch.Tensor) → torch.Tensor