Data Analysis and Machine Learning with Python

Explore the basics of pandas series and dataframes and master indexing and slicing to extract specific data. Use label-based (loc) and integer-based (iloc) indexing to access rows, columns, and subsets.

Explore how to sort, group, and aggregate data in pandas using a bookings dataframe, applying sort_values, groupby, and mean or min functions to compute destination-based averages.

Remove duplicates in a pandas data frame using the drop_duplicates function, optionally specifying a subset of columns to target specific fields and improve data accuracy.

Learn how to handle missing values in pandas with isnull, dropna, and fillna, encode categorical data using get_dummies, and normalize numerical features with MinMaxScaler for ready-to-model data.

Merge and join dataframes in pandas to combine accounts and transactions, using account ID as the key for merge and indices for join.

Explore handling date and time data in pandas by converting columns to datetime, switching time zones, resampling time series, and extracting year, month, and day.

Learn to use Pandas groupby to cluster a dataframe by game and name, then apply max, sum, and first on hours played to reveal patterns in top games and players.

Learn to create pivot tables in pandas using index, columns, and values to group data and summarize bookings by airline and destination.

Learn how to read and write data with pandas using the iris dataset, exporting to csv, excel, and json, and validating files in an export folder.

Apply pandas to compute summary statistics on the iris dataset, converting it to a dataframe and using describe, mean, median, mode, max, min, std, and variance to reveal data characteristics.

Explore plotting basics in Matplotlib with line, scatter, pie, and histogram plots. Visualize iris data relationships and tip proportions by day, and learn labeling, titles, and axis customization.

Explore Matplotlib subplots by creating a single figure with a pie chart of tip percentages by day and a green bar chart of tip counts from the Seaborn tips dataset.

Explore creating line, scatter, and bar plots in seaborn using the MPG, Texas, and Iris datasets; learn to set x and y axes, data, labels, and titles.

Practice pairplot, jointplot, and facetgrid in seaborn with the iris dataset to visualize pairwise relationships, distributions, and categorized plots.

Explore customizing seaborn plots by creating a scatter plot of total bill versus tip using the tips dataset, and adjust style, context, and labels for clear, poster-ready visuals.

Create scatter, bar, histogram, and line plots in Plotly using a Seaborn mpg dataset, then customize traces and layout to clearly visualize horsepower and mpg.

Explore three-dimensional data visualization with a 3d scatter plot in plotly, using horsepower, acceleration, and mpg from seaborn's mpg dataset, colored by origin, to create interactive insights.

Explore NumPy basics by creating arrays from lists, indexing and slicing, reshaping 1d arrays into 2d shapes, stacking arrays vertically or horizontally, and broadcasting operations on stock price data.

Explore numpy array creation and concatenation along the second axis, compute sums, and transpose arrays, and visualize distributions with matplotlib using histograms of uniform and normal data.

Explore exploratory data analysis to uncover structure and relationships in data through cleaning, pre-processing, summarization, feature engineering, using Python tools like pandas, seaborn, matplotlib, and numpy.

Explore an exploratory data analysis case study of concrete compressive strength, examining cement, blast furnace slag, fly ash, water, and superplasticizers, with heatmaps, scatter plots, and pair plots.

Gradient descent minimizes the linear regression cost function, the mean squared error, by iteratively updating slope and intercept to improve predictions. It highlights efficiency, learning rate, and potential drawbacks.

Learn how mean squared error loss (MSE) evaluates linear regression by measuring the average squared difference between true values and predictions, with Python numpy code and matplotlib plots.

Implement a one-variable linear regression in python with numpy and matplotlib, optimizing via gradient descent to predict y from x and visualize model fit.

Train a multiple linear regression model on synthetic data in Python using NumPy, visualize a 3D scatter plot of X1, X2, and Y, and track loss and gradient plots.

Explore how to apply linear regression to predict tips from total bill using scikit-learn, seaborn, and pandas. Build, visualize, and evaluate the model with train-test split, coefficients, intercept, and R^2.

Learn to acquire and clean World Bank API data with the WB data library and pandas, including GDP per capita and investment in education, via code demos.

Preprocess and clean data by merging two data frames on country from the World Bank API, renaming GDP and education columns, handling missing values, and normalizing with a minimax scaler.

Utilize the train test split function to divide data into training and testing sets, set test size and random state, and evaluate model performance on unseen data to avoid overfitting.

Train a linear regression model to explore how education investment relates to GDP per capita, then evaluate predictions with MAE, MSE, RMSE, and R-squared.

Visualize data with Matplotlib in Python to plot scatter plots and the linear regression line, revealing relationships between variables such as investment in education and GDP per capita.