Python's strength comes from its vast ecosystem of external libraries. Using pip (Python's package installer), you can add libraries for data science (Pandas, NumPy), visualization (Matplotlib, Seaborn), web development (Flask, Django), machine learning (Scikit-learn), and more. These libraries save time by providing pre-built solutions for common problems.
# Working with External Libraries
print("WORKING WITH EXTERNAL LIBRARIES")
print("=" * 60)
# Note: For actual use, install libraries first:
# pip install pandas numpy matplotlib seaborn
# Example 1: Introduction to Pandas (Data Analysis)
print("\n1. PANDAS - DATA ANALYSIS LIBRARY")
print("-" * 30)
print("Pandas is like Excel for Python - it handles tables of data.")
print("Main structures: Series (1D) and DataFrame (2D table)")
# Simulating Pandas functionality for demonstration
class MockDataFrame:
"""Mock DataFrame to demonstrate concepts without actual pandas"""
def __init__(self, data, columns=None):
self.data = data
self.columns = columns or [f"col{i}" for i in range(len(data[0]))]
def head(self, n=5):
"""Show first n rows"""
return self.data[:n]
def describe(self):
"""Show statistics"""
stats = {
"count": len(self.data),
"unique": len(set(row[0] for row in self.data)),
"example": self.data[0][0]
}
return stats
def filter_by(self, column, value):
"""Filter rows"""
col_index = self.columns.index(column)
return [row for row in self.data if row[col_index] == value]
# Sample data
student_data = [
["Alice", "Math", 85, "A"],
["Bob", "Science", 92, "A"],
["Charlie", "Math", 78, "B"],
["Diana", "Science", 88, "B"],
["Eve", "Math", 95, "A"],
["Frank", "Science", 82, "B"],
]
print("\nSample student data:")
df = MockDataFrame(student_data, ["Name", "Subject", "Score", "Grade"])
print("First 3 rows:")
for row in df.head(3):
print(f" {row}")
print(f"\nDataFrame stats: {df.describe()}")
print("\nFilter Math students:")
math_students = df.filter_by("Subject", "Math")
for student in math_students:
print(f" {student}")
# Example 2: NumPy - Numerical Computing
print("\n\n2. NUMPY - NUMERICAL COMPUTING")
print("-" * 30)
print("NumPy handles arrays and mathematical operations efficiently.")
print("Key features: Multi-dimensional arrays, mathematical functions, linear algebra")
# Simulating NumPy concepts
class MockArray:
"""Mock NumPy array"""
def __init__(self, data):
self.data = data
def shape(self):
if isinstance(self.data[0], list):
return (len(self.data), len(self.data[0]))
return (len(self.data),)
def mean(self):
return sum(self.data) / len(self.data) if self.data else 0
def reshape(self, new_shape):
"""Reshape array"""
flat_data = []
for item in self.data:
if isinstance(item, list):
flat_data.extend(item)
else:
flat_data.append(item)
# Simple reshape for 1D to 2D
rows, cols = new_shape
result = []
for i in range(rows):
row = flat_data[i*cols:(i+1)*cols]
result.append(row)
return MockArray(result)
# Demonstrate array operations
print("\nArray Operations:")
arr = MockArray([1, 2, 3, 4, 5, 6])
print(f"Array: {arr.data}")
print(f"Shape: {arr.shape()}")
print(f"Mean: {arr.mean():.2f}")
reshaped = arr.reshape((2, 3))
print(f"\nReshaped to 2x3: {reshaped.data}")
print(f"New shape: {reshaped.shape()}")
# Example 3: Matplotlib - Data Visualization
print("\n\n3. MATPLOTLIB - DATA VISUALIZATION")
print("-" * 30)
print("Matplotlib creates charts and graphs from data.")
print("Common plot types: Line, Bar, Scatter, Histogram, Pie")
# Simulating plot creation
class MockPlot:
"""Mock matplotlib plot"""
@staticmethod
def line_plot(x_data, y_data, title):
print(f"\nLine Plot: {title}")
print("X-axis:", x_data)
print("Y-axis:", y_data)
print("Creating line chart...")
# In real matplotlib: plt.plot(x_data, y_data); plt.title(title); plt.show()
@staticmethod
def bar_chart(labels, values, title):
print(f"\nBar Chart: {title}")
for label, value in zip(labels, values):
bar = "█" * (value // 2) # Simple bar representation
print(f"{label:10} {value:3} {bar}")
@staticmethod
def scatter_plot(x_data, y_data, title):
print(f"\nScatter Plot: {title}")
print(" X Y Plot")
for x, y in zip(x_data, y_data):
position = int((x + y) / 2)
plot_line = " " * position + "•"
print(f"{x:3} {y:3} {plot_line}")
# Create sample plots
months = ["Jan", "Feb", "Mar", "Apr", "May"]
sales = [150, 200, 175, 225, 300]
temperatures = [5, 7, 12, 15, 18, 22, 25, 23, 19, 14, 9, 6]
MockPlot.line_plot(months, sales, "Monthly Sales")
MockPlot.bar_chart(months, sales, "Monthly Sales Comparison")
# Scatter plot example
study_hours = [2, 3, 4, 5, 6, 7, 8]
exam_scores = [65, 70, 75, 80, 85, 90, 95]
MockPlot.scatter_plot(study_hours, exam_scores, "Study Hours vs Exam Scores")
# Example 4: Requests - HTTP Library (revisited with more depth)
print("\n\n4. REQUESTS - HTTP LIBRARY")
print("-" * 30)
print("Requests simplifies making HTTP requests.")
print("Common uses: API calls, web scraping, downloading files")
# Simulating requests functionality
class MockRequests:
"""Mock requests module"""
@staticmethod
def get(url, params=None, headers=None):
print(f"GET request to: {url}")
if params:
print(f"Parameters: {params}")
if headers:
print(f"Headers: {headers}")
# Return mock response
return MockResponse(200, "{\"data\": \"example response\"}")
@staticmethod
def post(url, data=None, json=None, headers=None):
print(f"POST request to: {url}")
if json:
print(f"JSON data: {json}")
# Return mock response
return MockResponse(201, "{\"id\": 123, \"status\": \"created\"}")
print("\nMaking API requests:")
# Mock API calls
response = MockRequests.get(
"https://api.example.com/data",
params={"page": 1, "limit": 10},
headers={"Authorization": "Bearer token123"}
)
print(f"Response status: {response.status_code}")
response = MockRequests.post(
"https://api.example.com/users",
json={"name": "John", "email": "john@example.com"}
)
print(f"Response status: {response.status_code}")
# Example 5: BeautifulSoup - HTML Parsing (revisited)
print("\n\n5. BEAUTIFULSOUP - HTML PARSING")
print("-" * 30)
print("BeautifulSoup parses HTML and XML documents.")
print("Common uses: Web scraping, data extraction from web pages")
# Simulating BeautifulSoup
class MockBeautifulSoup:
"""Mock BeautifulSoup"""
def __init__(self, html, parser):
self.html = html
def find(self, tag, attrs=None):
print(f"Finding <{tag}> with attributes: {attrs}")
return MockTag(f"<{tag}>Sample content")
def find_all(self, tag, attrs=None):
print(f"Finding all <{tag}> elements")
return [
MockTag(f"<{tag}>Item 1"),
MockTag(f"<{tag}>Item 2"),
MockTag(f"<{tag}>Item 3")
]
class MockTag:
def __init__(self, content):
self.content = content
def text(self):
return self.content.split(">")[1].split("<")[0]
def get(self, attr):
return f"value_of_{attr}"
print("\nParsing HTML:")
html_content = "Title
Content
"
soup = MockBeautifulSoup(html_content, "html.parser")
# Find elements
title = soup.find("h1")
print(f"Found title: {title.text()}")
all_paragraphs = soup.find_all("p")
print(f"Found {len(all_paragraphs)} paragraphs")
# Example 6: Pillow - Image Processing
print("\n\n6. PILLOW - IMAGE PROCESSING")
print("-" * 30)
print("Pillow (PIL) handles image manipulation.")
print("Common uses: Resizing, cropping, filtering, converting formats")
# Simulating Pillow functionality
class MockImage:
"""Mock PIL Image"""
def __init__(self, filename):
self.filename = filename
self.size = (800, 600) # Default size
self.format = "JPEG"
def resize(self, new_size):
print(f"Resizing image from {self.size} to {new_size}")
self.size = new_size
return self
def crop(self, box):
print(f"Cropping image with box: {box}")
return self
def save(self, new_filename, format=None):
format = format or self.format
print(f"Saving image as {new_filename} ({format})")
@staticmethod
def open(filename):
print(f"Opening image: {filename}")
return MockImage(filename)
print("\nImage processing example:")
img = MockImage.open("photo.jpg")
print(f"Original size: {img.size}")
# Resize image
img.resize((400, 300))
print(f"New size: {img.size}")
# Crop image
img.crop((100, 100, 300, 200))
# Save image
img.save("photo_resized.jpg", "JPEG")
# Example 7: Real-world project using multiple libraries
print("\n\n7. REAL-WORLD PROJECT: DATA ANALYSIS PIPELINE")
print("-" * 30)
print("Simulating a data analysis project with multiple libraries:")
print("1. Pandas for data manipulation")
print("2. NumPy for numerical operations")
print("3. Matplotlib for visualization")
print("4. Requests for data fetching")
class DataAnalysisPipeline:
"""Mock data analysis pipeline"""
def __init__(self):
self.data = None
def fetch_data(self, source):
"""Fetch data from source"""
print(f"Fetching data from {source}...")
# Simulate fetched data
self.data = [
{"Date": "2024-01-01", "Sales": 150, "Visitors": 200},
{"Date": "2024-01-02", "Sales": 180, "Visitors": 220},
{"Date": "2024-01-03", "Sales": 200, "Visitors": 250},
{"Date": "2024-01-04", "Sales": 175, "Visitors": 210},
{"Date": "2024-01-05", "Sales": 220, "Visitors": 280},
]
print(f"Fetched {len(self.data)} records")
def analyze_data(self):
"""Analyze the data"""
if not self.data:
print("No data to analyze")
return
print("\nData Analysis:")
print("=" * 40)
# Calculate statistics
total_sales = sum(item["Sales"] for item in self.data)
total_visitors = sum(item["Visitors"] for item in self.data)
avg_sales = total_sales / len(self.data)
conversion_rate = (total_sales / total_visitors) * 100
print(f"Total Sales: ${total_sales}")
print(f"Total Visitors: {total_visitors}")
print(f"Average Daily Sales: ${avg_sales:.2f}")
print(f"Conversion Rate: {conversion_rate:.1f}%")
# Find best day
best_day = max(self.data, key=lambda x: x["Sales"])
print(f"\nBest Day: {best_day['Date']}")
print(f"Sales: ${best_day['Sales']}, Visitors: {best_day['Visitors']}")
def visualize_data(self):
"""Create visualizations"""
if not self.data:
print("No data to visualize")
return
print("\nData Visualization:")
print("=" * 40)
# Extract data for plotting
dates = [item["Date"][-2:] for item in self.data] # Just day numbers
sales = [item["Sales"] for item in self.data]
visitors = [item["Visitors"] for item in self.data]
# Create simple text-based charts
print("\nSales Trend:")
max_sales = max(sales)
for date, sale in zip(dates, sales):
bar_length = int((sale / max_sales) * 20)
bar = "█" * bar_length
print(f"Day {date}: ${sale:3} {bar}")
print("\nVisitors Trend:")
max_visitors = max(visitors)
for date, visitor in zip(dates, visitors):
bar_length = int((visitor / max_visitors) * 20)
bar = "█" * bar_length
print(f"Day {date}: {visitor:3} visitors {bar}")
def generate_report(self):
"""Generate analysis report"""
print("\nGenerating Report:")
print("=" * 40)
report = f"""
DATA ANALYSIS REPORT
=====================
Period: January 1-5, 2024
Summary:
- Total Sales: ${sum(item['Sales'] for item in self.data)}
- Total Visitors: {sum(item['Visitors'] for item in self.data)}
- Average Conversion Rate: {(sum(item['Sales'] for item in self.data) / sum(item['Visitors'] for item in self.data)) * 100:.1f}%
Recommendations:
1. Focus marketing on high-conversion days
2. Analyze visitor demographics for better targeting
3. Consider promotions during low-traffic periods
Generated: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}
"""
print(report)
# Save report to file
with open("analysis_report.txt", "w") as f:
f.write(report)
print("Report saved to analysis_report.txt")
# Run the pipeline
print("\nRunning Data Analysis Pipeline:")
pipeline = DataAnalysisPipeline()
pipeline.fetch_data("sales_database")
pipeline.analyze_data()
pipeline.visualize_data()
pipeline.generate_report()
# Example 8: Installing and managing libraries
print("\n\n8. INSTALLING AND MANAGING LIBRARIES")
print("-" * 30)
print("Using pip (Python package installer):")
print("\nBasic commands:")
print("pip install package_name # Install a package")
print("pip install package_name==1.0.0 # Install specific version")
print("pip install -r requirements.txt # Install from requirements file")
print("pip uninstall package_name # Remove a package")
print("pip list # Show installed packages")
print("pip show package_name # Show package info")
print("\nExample requirements.txt file:")
requirements = """
# Project dependencies
pandas==1.5.3
numpy==1.24.3
matplotlib==3.7.1
requests==2.28.2
beautifulsoup4==4.12.2
pillow==9.5.0
"""
print(requirements)
print("\nVirtual environments (venv):")
print("python -m venv myenv # Create virtual environment")
print("source myenv/bin/activate # Activate (Linux/Mac)")
print("myenv\\Scripts\\activate # Activate (Windows)")
print("deactivate # Deactivate")
# Example 9: Popular Python libraries by category
print("\n\n9. POPULAR PYTHON LIBRARIES")
print("-" * 30)
libraries_by_category = {
"Data Science & Analysis": [
"pandas", "numpy", "scipy", "statsmodels"
],
"Data Visualization": [
"matplotlib", "seaborn", "plotly", "bokeh"
],
"Machine Learning": [
"scikit-learn", "tensorflow", "pytorch", "keras"
],
"Web Development": [
"django", "flask", "fastapi", "requests"
],
"Web Scraping": [
"beautifulsoup4", "scrapy", "selenium"
],
"Database": [
"sqlalchemy", "psycopg2", "pymongo"
],
"Testing": [
"pytest", "unittest", "nose"
],
"Automation": [
"selenium", "pyautogui", "schedule"
]
}
for category, libs in libraries_by_category.items():
print(f"\n{category}:")
for lib in libs:
print(f" • {lib}")
# Example 10: Creating your own library
print("\n\n10. CREATING YOUR OWN LIBRARY")
print("-" * 30)
print("Structure of a Python library:")
print("""
my_library/
README.md
setup.py
my_library/
__init__.py
module1.py
module2.py
tests/
test_module1.py
examples/
example_usage.py
""")
print("\nExample setup.py:")
setup_py = """
from setuptools import setup, find_packages
setup(
name="my_library",
version="0.1.0",
author="Your Name",
author_email="your.email@example.com",
description="A short description of your library",
packages=find_packages(),
install_requires=[
'requests>=2.25.0',
'pandas>=1.3.0',
],
python_requires='>=3.7',
)
"""
print(setup_py)
print("\nPublishing to PyPI:")
print("1. Create accounts on test.pypi.org and pypi.org")
print("2. Build package: python setup.py sdist bdist_wheel")
print("3. Upload: twine upload dist/*")
print("\nUsing your library:")
print("pip install my_library")
print("# Then in Python:")
print("import my_library")
print("from my_library import module1")