def build_model(): l_in = nn.layers.InputLayer((None, n_candidates_per_patient, 1,) + p_transform['patch_size']) l_in_rshp = nn.layers.ReshapeLayer(l_in, (-1, 1,) + p_transform['patch_size']) l_target = nn.layers.InputLayer((batch_size,)) penultimate_layer = load_pretrained_model(l_in_rshp) l = drop(penultimate_layer, name='drop_final') l = dense(l, 128, name='dense_final') l = nn.layers.DenseLayer(l, num_units=1, W=nn.init.Orthogonal(), nonlinearity=None, name='dense_p_benign') l = nn.layers.ReshapeLayer(l, (-1, n_candidates_per_patient, 1), name='reshape2patients') l_out = nn_lung.AggAllBenignExp(l, name='aggregate_all_nodules_benign') return namedtuple('Model', ['l_in', 'l_out', 'l_target'])(l_in, l_out, l_target)
def build_model(): l_in = nn.layers.InputLayer(( None, n_candidates_per_patient, 1, ) + p_transform['patch_size']) l_in_rshp = nn.layers.ReshapeLayer(l_in, ( -1, 1, ) + p_transform['patch_size']) l_target = nn.layers.InputLayer((batch_size, )) l = dnn.Conv3DDNNLayer(l_in_rshp, filter_size=3, num_filters=64, pad='valid', W=nn.init.Orthogonal(), nonlinearity=nn.nonlinearities.very_leaky_rectify) l = inrn_v2_red(l) l = inrn_v2(l) l = inrn_v2_red(l) l = inrn_v2(l) l = inrn_v2_red(l) l = inrn_v2(l) l = inrn_v2_red(l) l = nn.layers.GlobalPoolLayer(l) l = nn.layers.DenseLayer(l, num_units=1, W=nn.init.Orthogonal(), nonlinearity=None) l = nn.layers.ReshapeLayer(l, (-1, n_candidates_per_patient, 1)) l_out = nn_lung.AggAllBenignExp(l) return namedtuple('Model', ['l_in', 'l_out', 'l_target'])(l_in, l_out, l_target)
def build_model(): l_in = nn.layers.InputLayer(( None, n_candidates_per_patient, 1, ) + p_transform['patch_size']) l_in_un = nn_lung.Unbroadcast(l_in) l_in_rshp = nn.layers.ReshapeLayer(l_in_un, ( -1, 1, ) + p_transform['patch_size']) l_target = nn.layers.InputLayer((batch_size, )) l = conv3(l_in_rshp, num_filters=128) l = conv3(l, num_filters=128) l = max_pool(l) l = conv3(l, num_filters=128) l = conv3(l, num_filters=128) l = max_pool(l) l = conv3(l, num_filters=256) l = conv3(l, num_filters=256) l = conv3(l, num_filters=256) l = dense_prelu_layer(l, num_units=512) l = dense_prelu_layer(l, num_units=512) l = nn.layers.DenseLayer(l, num_units=1, W=nn.init.Orthogonal(), nonlinearity=None) l = nn.layers.ReshapeLayer(l, (-1, n_candidates_per_patient, 1)) l_out = nn_lung.AggAllBenignExp(l) return namedtuple('Model', ['l_in', 'l_out', 'l_target'])(l_in, l_out, l_target)