Source code for torchctr.datasets.avazu

#!/usr/bin/env python
# encoding: utf-8

import numpy as np
import pandas as pd
from .base import BaseDataset


class Avazu(BaseDataset):
    def __init__(self):
        super().__init__()

    def build_data(self):
        data = pd.read_csv("~/.torchctr/avazu/train_mini.data", engine="python")
        self.data = data[data.columns[1:]]
        self.y_column = "click"
        self.x_columns = [c for c in self.data.columns if c != self.y_column]

        # process x
        self.preprocess_x()
        # build feature
        self.preprocess_y()

        self.x = self.data[self.x_columns].to_numpy().astype(np.int)
        self.y = self.data[self.y_column].to_numpy().astype(np.float32)

        self.feature_dims = np.max(self.x, axis=0)