train_tools.models.MEDIARFormer =============================== .. py:module:: train_tools.models.MEDIARFormer Module Contents --------------- .. py:class:: MEDIARFormer(encoder_name='mit_b5', encoder_weights='imagenet', decoder_channels=(1024, 512, 256, 128, 64), decoder_pab_channels=256, in_channels=3, classes=3) Bases: :py:obj:`segmentation_models_pytorch.MAnet` MEDIAR-Former Model .. py:attribute:: segmentation_head :value: None .. py:attribute:: cellprob_head .. py:attribute:: gradflow_head .. py:method:: forward(x) Forward pass through the network