23 - Classes practice (recreating simple pandas DataFrame)

import numpy as np
from IPython.display import display, HTML

Recreating the DataFrame class

class DataFrameCopy:
    def __init__(self, data: dict):
        self.data = data
        self.create_index() # Run method to create index on initializing of object

    # Add index to dataset
    def create_index(self):
        df = {} 
        max_rows = 0
       
        for key, value in self.data.items():  # return max number of rows to propagate index
            if len(value) > max_rows:
                max_rows = len(value)
        
        index_length = list(range(0, max_rows)) # Create index column
        df['index'] = index_length # Add index to temp dict
        df.update(self.data) # Add initialize data to temp dict and reassign data that includes index
        self.data = df

    # Replicating the head function of pandas
    def head(self, rows: int):
        table = '<table style="border-collapse: collapse; font-family: Arial; font-size: 14px;">'

        # loop to insert columns
        table += '<tr>'
        for key in self.data.keys():
            table += f'<th>{key}</th>'
        table += '</tr>'

        # Below is the loop to insert rows values for each column
        i = 0 # i is used to keep track of the row level, whereby
        while True:
            for key, value in self.data.items():
                table += f"<td>{value[i]}</td>"
            table += "</tr>"
            i += 1

            # When desired amount of rows have been inserted in html, break loop
            if i == rows:
               break
        table += "</table>"

        # Read string as html
        chart = HTML(table)
        
        return display(chart)

Initializing DataFrameCopy class with data in the form of dictionary just like in pandas

data = {'Col 1': [1, 2, 3, 4, 5], 'Col 2': [6, 7, 8, 9, 10]}
df = DataFrameCopy(data)

Below the data from the object is called where we can see that the index has been automatically created

df.data

{'index': [0, 1, 2, 3, 4], 'Col 1': [1, 2, 3, 4, 5], 'Col 2': [6, 7, 8, 9, 10]}

Using the head method for the class works simply like the pandas version, where we can choose the amount of rows we would like to show. The method then displays a html table of the data for improved readability.

df.head(4)

index	Col 1	Col 2
0	1	6
1	2	7
2	3	8
3	4	9