"""
Tabular2 uni-modal model.
"""
import torch.nn as nn
from fusilli.fusionmodels.base_model import ParentFusionModel
from fusilli.utils import check_model_validity
[docs]
class Tabular2Unimodal(ParentFusionModel, nn.Module):
"""Tabular2 uni-modal model.
This class implements a uni-modal model using only the 2nd type of tabular data.
Attributes
----------
mod2_layers : nn.ModuleDict
Dictionary containing the layers of the 2nd type of tabular data.
fused_dim : int
Dimension of the fused layer.
fused_layers : nn.Sequential
Sequential layer containing the fused layers.
final_prediction : nn.Sequential
Sequential layer containing the final prediction layers.
fused_dim : int
Dimension of the fused layer.
"""
#: str: Name of the method.
method_name = "Tabular2 uni-modal"
#: str: Modality type.
modality_type = "tabular2"
#: str: Fusion type.
fusion_type = "unimodal"
[docs]
def __init__(self, prediction_task, data_dims, multiclass_dimensions):
"""
Parameters
----------
prediction_task : str
Type of prediction to be performed.
data_dims : list
List containing the dimensions of the data.
multiclass_dimensions : int
Number of classes in the multiclass classification task.
"""
ParentFusionModel.__init__(
self, prediction_task, data_dims, multiclass_dimensions
)
self.prediction_task = prediction_task
self.set_mod2_layers()
self.get_fused_dim()
self.set_fused_layers(self.fused_dim)
self.calc_fused_layers()
[docs]
def get_fused_dim(self):
"""
Get the number of features of the fused layers.
Returns
-------
None
"""
self.fused_dim = list(self.mod2_layers.values())[-1][0].out_features
[docs]
def calc_fused_layers(self):
"""
Calculates the fused layers.
Returns
-------
None
"""
check_model_validity.check_dtype(self.mod2_layers, nn.ModuleDict, "mod2_layers")
# check fused layers
self.get_fused_dim()
self.fused_layers, out_dim = check_model_validity.check_fused_layers(
self.fused_layers, self.fused_dim
)
self.set_final_pred_layers(out_dim)
[docs]
def forward(self, x):
"""
Forward pass of the model.
Parameters
----------
x : torch.Tensor
Input tensor.
Returns
-------
torch.Tensor
Output tensor.
"""
check_model_validity.check_model_input(x)
x_tab2 = x
for layer in self.mod2_layers.values():
x_tab2 = layer(x_tab2)
out_fuse = self.fused_layers(x_tab2)
out_pred = self.final_prediction(out_fuse)
return out_pred