Inheritance diagram for PriorModel:

Public Member Functions
def	__init__ (self, int n_output)

torch.tensor	forward (self, torch.tensor x)

Public Attributes
	hidden1
	Linear layer with 9 inputs and 128 outputs.

	act1
	ReLU activation layer.

	hidden2
	Batchnormalization layer.

	hidden3
	Linear layer with 128 inputs and 64 outputs.

	act2
	ReLU activation layer.

	hidden4
	Batchnormalization layer.

	hidden5
	Linear layer with 64 inputs and 32 outputs.

	act3
	ReLU activation layer.

	hidden6
	Batchnormalization layer.

	hidden7
	Linear layer with 32 inputs and outputs for each particle in the particlelist.

	act4
	Softmax activation layer.

Detailed Description

Pytorch model for PID prior probability calculation.

Attributes:
    hidden1: Linear layer with 9 inputs and 128 outputs.
    act1: An RELU activation layer.
    hidden2: A batch normalization layer.
    hidden3: Linear layer with 128 inputs and 64 outputs.
    act2: An RELU activation layer.
    hidden4: A batch normalization layer.
    hidden5: Linear layer with 64 inputs and 32 outputs.
    act3: An RELU activation layer.
    hidden6: A batch normalization layer.
    hidden7: Linear layer with 9 inputs and 128 outputs.
    act4: A softmax activation layer.

Definition at line 105 of file priorDataLoaderAndModel.py.

Constructor & Destructor Documentation

◆ init()

def __init__	(		self,
		int	n_output
	)

Initialize the PID prior probability model.

Parameter:
    n_output (int): Number of output nodes.

Definition at line 124 of file priorDataLoaderAndModel.py.

    def __init__(self, n_output: int):
        """
        Initialize the PID prior probability model.
 
        Parameter:
            n_output (int): Number of output nodes.
 
        """
        super().__init__()
        
        self.hidden1 = Linear(9, 128)
        kaiming_uniform_(self.hidden1.weight, nonlinearity="relu")
        
        self.act1 = ReLU()
        
        self.hidden2 = BatchNorm1d(128)
        
        self.hidden3 = Linear(128, 64)
        kaiming_uniform_(self.hidden3.weight, nonlinearity="relu")
        
        self.act2 = ReLU()
        
        self.hidden4 = BatchNorm1d(64)
        
        self.hidden5 = Linear(64, 32)
        kaiming_uniform_(self.hidden5.weight, nonlinearity="relu")
        
        self.act3 = ReLU()
        
        self.hidden6 = BatchNorm1d(32)
        
        self.hidden7 = Linear(32, n_output)
        xavier_uniform_(self.hidden7.weight)
        
        self.act4 = Softmax(dim=1)
 

Member Function Documentation

◆ forward()

torch.tensor forward	(		self,
		torch.tensor	x
	)

Gives PID prior probabilities for the input features.

Parameter:
    x (torch.tensor): A 2D tensor containing features for a particle as a row.

Returns:
    A torch tensor containing PID prior probabilities for the provided features.

Definition at line 160 of file priorDataLoaderAndModel.py.

    def forward(self, x: torch.tensor) -> torch.tensor:
        """
        Gives PID prior probabilities for the input features.
 
        Parameter:
            x (torch.tensor): A 2D tensor containing features for a particle as a row.
 
        Returns:
            A torch tensor containing PID prior probabilities for the provided features.
        """
        x = self.hidden1(x)
        x = self.act1(x)
        x = self.hidden2(x)
        x = self.hidden3(x)
        x = self.act2(x)
        x = self.hidden4(x)
        x = self.hidden5(x)
        x = self.act3(x)
        x = self.hidden6(x)
        x = self.hidden7(x)
        x = self.act4(x)
        return x