Python Libraries

PYTHON TOOLBOX

Same data. Different tools. Choose a library and see exactly what role it plays in the machine-learning workflow.

DataLoad & clean

→

PrepareArrays

→

ModelLearn

→

VisualizeUnderstand

▤

titanic.csv891 rows × 12 columns

→

pandas DataFramepandas

name	age	sex	survived
Braund	22	male	0
Cumings	38	female	1
Heikkinen	NaN	female	1
Futrelle	35	female	0
Allen	4	male	1

pandas turns raw tabular data into a structured DataFrame that is easy to inspect and clean.

▦

Data loading & cleaning

ROLELoad & Clean

Example

import pandas as pd

df = pd.read_csv("titanic.csv")

df["age"] = df["age"].fillna(
  df["age"].median()
)

df.head()

💡

REMEMBERThink: rows + columns of data

▦

pandas

CSV, Excel & tabular data

◇

NumPy

Fast numerical operations

⌘

scikit-learn

Regression & classification

◉

PyTorch

Building neural networks

▥

Matplotlib

Visualizing data & results