ML - Lab 1 - Basics in DataFrame

from random import randrange
import pandas as pd

# Create a DataFrame
nums = [1, 4, 9, 16, 25, 36, 49]
strs = ["aa", "bb", "cc", "dd", "ee", "ff", "gg"]
bools = [True, False, True, True, False, False, True]
df = pd.DataFrame(list(zip(nums, strs, bools)), columns=['ns', 'ss', 'bs'])
print("**********************************")
print(df)
print()
print(df.ns)
print()
print(df['ss'])
print()

print("**********************************")
# Rows
row_ind = 0
row = df.loc[row_ind]
print(str(row) + ", length = " + str(len(row)))
print()
# Columns
col_name = 'bs'
col = df.loc[:, col_name]
print(str(col) + ", length = " + str(len(col)))
print()
# Elements
element = df.loc[1, "bs"]
print(element)
print()
# Columns with condition True
rows_ind = df.loc[df.bs]
print(rows_ind)
print()
print()

# Load a file .csv
print("**********************************")
people = pd.read_csv("people.txt")
print(people)
print()
people = pd.read_csv("people2.txt", sep=";", header=None, names=["Age", "Height", "Weight"])
print(people)
print()
# Missing values are found by average
people = people.fillna(people.mean())
print(people)
print()
print()

# Statistical info
print("**********************************")
print("Head: ")
print(people.head())
print()
print("Describe: ")
print(people.describe())
print()
print("Null elements in each column: ")
print(people.isnull().sum())
print()
print()

# Convert categorial data to numerical
from sklearn.preprocessing import OneHotEncoder
# Create Encoder
encoder = OneHotEncoder(handle_unknown="ignore", sparse=False)
encoder.fit(people)
people_one_hot = encoder.transform(people)
print("**********************************")
print(people_one_hot)