PriorDataLoader Class Reference

Inheritance diagram for PriorDataLoader:

Public Member Functions
def	__init__ (self, str path, str key, list particlelist, list labels)
def	__getitem__ (self, index)
def	__len__ (self)
torch.tensor	get_split (self, float n_test=0.1)

Public Attributes
	x
	The tensor of features.
	y
	The tensor of labels.

Detailed Description

Dataloader for PID prior probability training.

Attributes:
    x (np.array): Array containing feature data with a second order combination of momentum, cos(theta) and transverse momentum.
    y (np.array): Array containing the label encoded PDG values.

Definition at line 26 of file priorDataLoaderAndModel.py.

Constructor & Destructor Documentation

__init__()

def __init__	(		self,
		str	path,
		str	key,
		list	particlelist,
		list	labels
	)

Initialize the dataloader for PID prior training.

Parameters:
    path (str): Path to the root file containing the data.
    key (str): Key (i.e. path) of the tree within the root file.
    particlelist (list(int)): List of particle PDG values for which the model has to be trained.
    labels (str): Labels of pandas columns containing cos(theta), momentum and PDG values (in this order).

Definition at line 36 of file priorDataLoaderAndModel.py.

    def __init__(self, path: str, key: str, particlelist: list, labels: list):
        """
        Initialize the dataloader for PID prior training.
 
        Parameters:
            path (str): Path to the root file containing the data.
            key (str): Key (i.e. path) of the tree within the root file.
            particlelist (list(int)): List of particle PDG values for which the model has to be trained.
            labels (str): Labels of pandas columns containing cos(theta), momentum and PDG values (in this order).
 
        """
        data = ur.open(path)
        data = data[key].pandas.df(labels)
        df = data.dropna().reset_index(drop=True)
        df.loc[:, labels[2]] = df.loc[:, labels[2]].abs()
        droplist = np.setdiff1d(np.unique(df[labels[2]].values), particlelist)
        for i in droplist:
            df = df.drop(df.loc[df[labels[2]] == i].index).reset_index(drop=True)
        x = df.values[:, 0:2]
        x = np.hstack((x, (np.sin(np.arccos(x[:, 0])) * x[:, 1]).reshape(-1, 1)))
        pol = PolynomialFeatures(2, include_bias=False)
        x = pol.fit_transform(x)
        
        self.x = x.astype("float32")
        y = df.values[:, 2]
        le = LabelEncoder()
        y = le.fit_transform(y)
        
        self.y = y.astype("int64")
 

Member Function Documentation

__getitem__()

def __getitem__	(		self,
			index
	)

Function to get feature and label tensors at the given index location.

Parameters:
    index (int): The index of required tensors.

Returns:
    Tensors of features and labels at the given index.

Definition at line 66 of file priorDataLoaderAndModel.py.

    def __getitem__(self, index):
        """
        Function to get feature and label tensors at the given index location.
 
        Parameters:
            index (int): The index of required tensors.
 
        Returns:
            Tensors of features and labels at the given index.
        """
        return [self.x[index], self.y[index]]
 

__len__()

def __len__

(

self

)

Function to obtain length of a tensor.

Parameters:
    None.

Returns:
    Number of feature sets.

Definition at line 78 of file priorDataLoaderAndModel.py.

    def __len__(self):
        """
        Function to obtain length of a tensor.
 
        Parameters:
            None.
 
        Returns:
            Number of feature sets.
        """
        return len(self.x)
 

get_split()

torch.tensor get_split	(		self,
		float	n_test = 0.1
	)

Split the input data into training and validation set.

Parameter:
    n_test (float): Ratio of number of particles to be taken in the validation set to that of training set.

Return:
    A randomly split data set with the ratio given by 'n_test'.

Definition at line 90 of file priorDataLoaderAndModel.py.

    def get_split(self, n_test: float = 0.1) -> torch.tensor:
        """
        Split the input data into training and validation set.
 
        Parameter:
            n_test (float): Ratio of number of particles to be taken in the validation set to that of training set.
 
        Return:
            A randomly split data set with the ratio given by 'n_test'.
        """
        test_size = round(n_test * len(self.x))
        train_size = len(self.x) - test_size
        return random_split(self, [train_size, test_size])
 
 

Member Data Documentation

x

x

The tensor of features.

Definition at line 59 of file priorDataLoaderAndModel.py.

y

y

The tensor of labels.

Definition at line 64 of file priorDataLoaderAndModel.py.

---

The documentation for this class was generated from the following file:

• analysis/scripts/priorDataLoaderAndModel.py