3x3 convolution with padding

3x3 convolution with padding

3x3 convolution with padding

3x3 convolution with padding

3x3 convolution with padding

3x3 convolution with padding

3x3 convolution with padding

Initialize this class from some cli arguments. Used in train, test.

Calculate normalized generalized balanced weights

Calculate normalized generalized balanced weights

Calculate normalized generalized balanced weights

Calculate normalized generalized balanced weights

Get the per-example sampling weights return shape: [capacity]

Get the per-example sampling weights return shape: [capacity]

Get the per-example sampling weights return shape: [capacity]

Get the per-example sampling weights return shape: [capacity]

Get the per-example sampling weights return shape: [capacity]

Get the per-example sampling weights return shape: [capacity]

seperate backbone optimizer and classifier optimizer by Kaihua

seperate backbone optimizer and classifier optimizer by Kaihua

Reset the manually designed weights and update the whole tree accordingly. @rescale: rescale the fixed_weights such that

Reset the manually designed weights and update the whole tree accordingly. @rescale: rescale the fixed_weights such that

Reset the manually designed weights and update the whole tree accordingly. @rescale: rescale the fixed_weights such that

Reset the manually designed weights and update the whole tree accordingly. @rescale: rescale the fixed_weights such that

Full training logic

Update the whole tree based on per-example sampling weights

Update the whole tree based on per-example sampling weights

Access a nested object in tree by sequence of keys.

Load from cRT pretrain :param cRT pretrain path to the checkpoint of cRT pretrain

setup GPU device if available, move model into configured device

Resume from saved checkpoints :param resume_path: Checkpoint path to be resumed

Set a value in a nested object in tree by sequence of keys.

Training logic for an epoch :param epoch: Current epoch number

Validate after training an epoch :param epoch: Integer, current training epoch. :return: A log that contains information abo