(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, idx)

(self, data_info, data_cfg, mode='testing')

(self, t=0.0)

(self, args, cfgs, local_rank, global_rank, logger, tb_writer, model)

(self)

(self, resblock=True, input_channel=3, encoder_ratio=16, decoder_ratio=16)

(self, num_features=8)

(self)

(self)

(self, in_planes, planes, norm_fn='group', stride=1)

(self, output_dim=None, norm_fn='batch', downsample=3)

(self, init_fmap1, init_fmap2, geo_volume, num_levels=2, radius=4)

(self, corr_levels, corr_radius, volume_channel)

(self, input_dim=128, hidden_dim=256, output_dim=1)

(self, n_gru_layers, corr_levels, corr_radius, volume_channel, hidden_dims=None)

(self, in_channels, backbone_channels=None)

(self)

(self)

(self, in_channels, out_channels, kernel_size, stride=1, padding=0, bias=False, norm_layer=No

(self, in_channels, out_channels, kernel_size, stride=1, padding=0, bias=False, norm_layer=No

(self, inp, oup, stride, expanse_ratio, dilation=1)

(self, dim, img_feat_dim)

(self, cfgs)

(self, backbone='MobileNetv2')

(self, args, cfgs, local_rank, global_rank, logger, tb_writer)

(self, args)

(self, args, cfgs, local_rank, global_rank, logger, tb_writer)

(self, threshold: int = 3, validation_max_disp: int = -1, loss_weight: list = None)

(self, left, right, disp=None, sampled_cols=None, sampled_rows=None, occ_mask=None, occ_mask_

(self)

(self, *args, **kwargs)

(self)

(self, num_blocks=8, feature_dim=16, expansion=3)

(self, num_pos_feats=64, temperature=10000, normalize=False, scale=None)

(self, cal: nn.Module, ot: bool = True)

(self, embed_dim, num_heads)

(self, hidden_dim: int, nhead: int)

(self, hidden_dim: int, nhead: int)

(self, num_layers, num_input_features, bn_size, growth_rate, drop_rate, memory_efficient=False)

(self, num_input_features, num_output_features)

(self, in_channels: int, out_channels: int, scale: int = 2)

(self, in_channels: int, out_channels: int)

(self, in_planes, planes, norm_fn='group', stride=1)

(self, in_planes, planes, norm_fn='group', stride=1)

(self, output_dim=128, norm_fn='batch', dropout=0.0, downsample=3)

(self, output_dim=[128], norm_fn='batch', dropout=0.0, downsample=3)

(self)

(self, in_channels, out_channels, deconv=False, is_3d=False, bn=True, relu=True, **kwargs)

(self, in_channels, out_channels, deconv=False, is_3d=False, IN=True, relu=True, **kwargs)

(self, in_channels, out_channels, deconv=False, is_3d=False, concat=True, keep_concat=True, IN=True, relu=True

(self, cv_chan, feat_chan)

(self, dims, mode='sintel', divis_by=8)

(self, geo_volume, num_levels=2, radius=4)

(self, in_channels)

(self, args)

(self, input_dim=128, hidden_dim=256, output_dim=2)

(self, input_dim=128, hidden_dim=128, output_dim=1)

(self, hidden_dim=128, input_dim=192+128)

(self, args)

(self, args, hidden_dim=96)

(self, args, cfgs, local_rank, global_rank, logger, tb_writer)

(self, cfgs)

(self, cv_chan, feat_chan)

(self, in_planes, planes, norm_fn='group', stride=1)

(self, output_dim=None, norm_fn='batch', downsample=3)

(self, init_fmap1, init_fmap2, geo_volume, num_levels=2, radius=4)

(self, corr_levels, corr_radius, volume_channel)

(self, input_dim=128, hidden_dim=256, output_dim=1)

(self, n_gru_layers, corr_levels, corr_radius, volume_channel, hidden_dims=None)

(self)

(self, args, cfgs, local_rank, global_rank, logger, tb_writer)

(self, in_channels, backbone_channels=None)

(self, cfgs)

(self, max_disparity, in_chan, hidden_chan, head=1, weighted=False)

(self, cv_chan, im_chan, D)

( self, max_disparity=192, matching_head=1, gce=True,

( self, max_disp=192, gce=True, matching_weighted=False,

(self, inplanes, planes, stride, downsample, pad, dilation)

(self, in_channels, out_channels, deconv=False, is_3d=False, bn=True, relu=True, **kwargs)

( self, inplanes: int, planes: int, stride: int = 1,

( self, in_planes: int, out_planes: int, kernel_size: int = 3,

( self, inp: int, oup: int, stride: int, expand_ra

(self, in_channels, out_channels, stride=1, bn=True, relu=True)