LEARN COMPLETE PYTHON IN 24 HOURS

10. Real-World Projects & Case Studies

These four hands-on projects apply everything you've learned: data loading, EDA, preprocessing, modeling, evaluation, visualization, and interpretation. They are designed to be portfolio-ready and commonly asked about in interviews.

10.1 Project 1: House Price Prediction (Regression)

Goal: Predict house prices based on features (classic regression problem).

Dataset: California Housing (built-in in sklearn) or use Kaggle's House Prices dataset.

Steps & Code

Python

import pandas as pd import numpy as np from sklearn.datasets import fetch_california_housing from sklearn.model_selection import train_test_split from sklearn.preprocessing import StandardScaler from sklearn.ensemble import RandomForestRegressor from sklearn.metrics import mean_squared_error, r2_score import matplotlib.pyplot as plt import seaborn as sns # 1. Load data housing = fetch_california_housing(as_frame=True) df = housing.frame X = df.drop("MedHouseVal", axis=1) y = df["MedHouseVal"] # 2. EDA (quick look) print(df.describe()) sns.heatmap(df.corr(), annot=True, cmap='coolwarm', fmt='.2f') plt.title("Correlation Matrix – House Prices") plt.show() # 3. Preprocessing X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42) scaler = StandardScaler() X_train_scaled = scaler.fit_transform(X_train) X_test_scaled = scaler.transform(X_test) # 4. Model training & evaluation model = RandomForestRegressor(n_estimators=100, random_state=42) model.fit(X_train_scaled, y_train) predictions = model.predict(X_test_scaled) rmse = np.sqrt(mean_squared_error(y_test, predictions)) r2 = r2_score(y_test, predictions) print(f"RMSE: {rmse:.3f} (lower is better)") print(f"R² Score: {r2:.3f} (closer to 1 is better)") # 5. Feature importance importances = pd.Series(model.feature_importances_, index=X.columns) importances.sort_values(ascending=False).plot(kind='bar') plt.title("Feature Importance – House Price Prediction") plt.show()

Key Takeaways:

• Median Income is usually the strongest predictor

• RMSE in range 0.45–0.55 is good for this dataset

• Try XGBoost or LightGBM for better performance

Improvements:

• Add feature engineering (rooms per household, age buckets)

• Hyperparameter tuning (GridSearchCV)

• Deploy as Streamlit app

10.2 Project 2: Customer Churn Prediction (Classification)

Goal: Predict whether a customer will leave (churn) — imbalanced classification problem.

Dataset: Telco Customer Churn (Kaggle or use seaborn example)

Steps & Code

Python

import pandas as pd from sklearn.model_selection import train_test_split from sklearn.preprocessing import StandardScaler, OneHotEncoder from sklearn.compose import ColumnTransformer from sklearn.pipeline import Pipeline from sklearn.ensemble import RandomForestClassifier from sklearn.metrics import classification_report, roc_auc_score import seaborn as sns # 1. Load & quick EDA df = pd.read_csv("https://raw.githubusercontent.com/IBM/telco-customer-churn-on-icp4d/master/data/Telco-Customer-Churn.csv") print(df['Churn'].value_counts(normalize=True)) # imbalanced ~73% No # 2. Preprocessing df = df.drop(['customerID'], axis=1) df['TotalCharges'] = pd.to_numeric(df['TotalCharges'], errors='coerce') df = df.dropna() X = df.drop('Churn', axis=1) y = df['Churn'].map({'Yes': 1, 'No': 0}) categorical = X.select_dtypes(include='object').columns numeric = X.select_dtypes(include=['int64', 'float64']).columns preprocessor = ColumnTransformer( transformers=[ ('num', StandardScaler(), numeric), ('cat', OneHotEncoder(drop='first', handle_unknown='ignore'), categorical) ]) X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y, random_state=42) # 3. Model pipeline pipeline = Pipeline([ ('preprocessor', preprocessor), ('classifier', RandomForestClassifier(class_weight='balanced', random_state=42)) ]) pipeline.fit(X_train, y_train) y_pred = pipeline.predict(X_test) y_prob = pipeline.predict_proba(X_test)[:, 1] print(classification_report(y_test, y_pred)) print("ROC-AUC:", roc_auc_score(y_test, y_prob)) # 4. Confusion matrix visualization sns.heatmap(pd.crosstab(y_test, y_pred), annot=True, fmt='d', cmap='Blues') plt.title("Confusion Matrix – Churn Prediction") plt.xlabel("Predicted") plt.ylabel("Actual") plt.show()

Key Takeaways:

• Class imbalance → use class_weight='balanced' or SMOTE

• Focus on Recall (catching churners) and ROC-AUC

• Top features: Contract type, tenure, monthly charges

Improvements:

• Try XGBoost / LightGBM

• Add SMOTE in pipeline

• Create dashboard with Streamlit

10.3 Project 3: Sentiment Analysis on Reviews (NLP)

Goal: Classify product/movie reviews as positive/negative/neutral.

Dataset: Amazon Reviews or IMDb (use Hugging Face datasets)

Easy & powerful method: Hugging Face Transformers

Python

from transformers import pipeline import pandas as pd # Load sentiment pipeline (pre-trained model) sentiment = pipeline("sentiment-analysis", model="nlptown/bert-base-multilingual-uncased-sentiment") # Sample reviews reviews = [ "This phone is amazing! Battery lasts all day.", "Worst product ever. Broke in 2 days.", "It's okay, nothing special but works fine.", "Absolutely love it! Best purchase this year." ] results = sentiment(reviews) for review, res in zip(reviews, results): print(f"Review: {review}") print(f"Sentiment: {res['label']} (score: {res['score']:.4f})\n")

Custom model with scikit-learn + TF-IDF

Python

from sklearn.feature_extraction.text import TfidfVectorizer from sklearn.linear_model import LogisticRegression from sklearn.model_selection import train_test_split from sklearn.metrics import classification_report # Assume df has 'review' and 'sentiment' columns (1=positive, 0=negative) X = df['review'] y = df['sentiment'] X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42) vectorizer = TfidfVectorizer(max_features=5000, stop_words='english') X_train_vec = vectorizer.fit_transform(X_train) X_test_vec = vectorizer.transform(X_test) model = LogisticRegression(max_iter=1000) model.fit(X_train_vec, y_train) y_pred = model.predict(X_test_vec) print(classification_report(y_test, y_pred))

Key Takeaways:

• Pre-trained transformers (Hugging Face) → best accuracy with almost no code

• TF-IDF + Logistic Regression → fast baseline, good interpretability

Improvements:

• Fine-tune BERT/RoBERTa

• Add emoji/text cleaning

• Create Streamlit app for live prediction

10.4 Project 4: Sales Dashboard & EDA Report

Goal: Create an interactive EDA & sales dashboard using Streamlit.

Install

Bash

pip install streamlit pandas plotly

Full code (save as app.py)

Python

import streamlit as st import pandas as pd import plotly.express as px st.title("Sales Dashboard & EDA Report") # Upload data uploaded_file = st.file_uploader("Upload your sales CSV", type="csv") if uploaded_file: df = pd.read_csv(uploaded_file) st.subheader("Data Overview") st.dataframe(df.head()) st.subheader("Summary Statistics") st.write(df.describe()) # Interactive filters category = st.selectbox("Select Category", df.columns) # Visualizations fig1 = px.histogram(df, x=category, title=f"Distribution of {category}") st.plotly_chart(fig1) fig2 = px.box(df, x="Region", y="Sales", title="Sales by Region") st.plotly_chart(fig2) st.subheader("Top Products") top_products = df.groupby("Product")["Sales"].sum().nlargest(10) st.bar_chart(top_products)

Run

Bash

streamlit run app.py

Key Takeaways:

• Streamlit = fastest way to turn data scripts into interactive dashboards

• Plotly = interactive charts (zoom, hover)

• Great for EDA reports, stakeholder presentations

This completes the full Real-World Projects & Case Studies section — now you have four portfolio-ready projects!

📚 Amazon Book Library

All my books are FREE on Amazon Kindle Unlimited🌍 Exclusive Country-Wise Amazon Book Library – Only Here!

On GlobalCodeMaster.com you’ll find complete, ready-to-use lists of my books with direct Amazon links for every country.

Belong to India, Australia, USA, UK, Canada or any other country? Just click your country’s link and enjoy:

✅ Any eBook FREE on Kindle Unlimited ✅ Or buy at incredibly low prices

400+ fresh books written in 2025-2026 with today’s latest AI, Python, Machine Learning & tech trends – nowhere else will you find this complete country-wise collection on one platform!

Choose your country below and start reading instantly 🚀

BOOK LIBRARY USA 2026 LINK

BOOK LIBRARY INDIA 2026 LINK

BOOK LIBRARY AUSTRALIA 2026 LINK

BOOK LIBRARY CANADA 2026 LINK

BOOK LIBRARY UNITED KINGDOM 2026 LINK

BOOK LIBRARY GERMANY 2026 LINK

BOOK LIBRARY FRANCE 2026 LINK

BOOK LIBRARY ITALY 2026 LINK

BOOK LIBRARY SPAIN 2026 LINK

BOOK LIBRARY NETHERLANDS 2026 LINK

BOOK LIBRARY BRAZIL 2026 LINK

BOOK LIBRARY MEXICO 2026 LINK

BOOK LIBRARY JAPAN 2026 LINK

BOOK LIBRARY POLAND 2026 LINK

BOOK LIBRARY IRELAND 2026 LINK

BOOK LIBRARY SWEDEN 2026 LINK

BOOK LIBRARY BELGIUM 2026 LINK