Practical Machine Learning with Python

Master arithmetic operators in expressions, including addition, subtraction, multiplication, division, floor division, and the power operator, using variables like x and y.

Explore relational operators for comparing values, including greater than, less than, greater than or equal, less than or equal, and equal to or not equal.

Explore logical and boolean operators, including and, or, and not, using true and false values to evaluate expressions and understand how truth tables guide decisions.

Explore bitwise operators and binary numbers, learn how 64-bit representation works, and use and, or, and shift left, and shift right to compute results and convert binary to decimal.

Learn how assignment operators work in Python, from = and += to -=, *=, /=, //=, %=, and explore power, bitwise and shift operators.

Explore identity operators such as is and is not, and membership operators such as in and not in, and observe how lists may share or hold different copies in memory.

Explore Python's math library for mathematical operations, including import math, exp for powers, log for natural and base-change, floor and ceiling, and the pi constant for readable expressions.

Master variables in Python by using assignment to set values and observe dynamic typing as types change automatically; learn to find a variable's data type.

Explore Python data types from booleans and integers to strings and complex numbers, with typing and type checks. See time, lists, dictionaries, sets, and how keys access values.

Explore typecasting in Python by converting between strings and numbers using int and floor conversions, recognizing when a value is compatible or results in errors, and casting on the fly.

Explore how booleans drive conditional logic by evaluating true and false values in if statements, including zero, none, and strings, with indentation affecting code blocks.

Explore strings in Python, learn to declare string variables, measure length with len, and access characters via zero-based indexing, including negative indexing and basic slicing to extract substrings.

Learn how to handle special characters in Python strings by escaping quotes and using backslashes. See how printing outputs prevents string termination errors and supports including quotes.

Master string handling by splitting strings with various separators, applying maxsplit limits, stripping edge whitespace, and joining strings through concatenation, while understanding string mutability.

Learn how lists work in Python by using square brackets, checking length with len, and accessing or mutating elements with zero-based and negative indexing, slices, and append.

Explore Python list slicing, including start and end bounds, negative indices, and how the end is excluded. Practice reversing order and selecting every nth element to extract patterns.

Explore different kinds of lists in Python, including numerical, floats, strings, and nested and mixed lists, and learn how to create and combine them.

Split a string by spaces to create a list of elements, then join the list with a space using Python's join method to reconstruct the original string.

Learn how to concatenate two lists using the plus operator, creating a new list, and assign it to another variable to preserve the originals.

Learn to create dictionaries with key-value pairs, fetch by key, and update values. Use methods like keys, values, and items, and practice deleting pairs, clearing, or deleting the dictionary.

Explore a dictionary and its methods, using len to measure size, get to retrieve values, test key existence, and identify min/max values with their corresponding keys.

Learn how to work with nested dictionaries by accessing values with keys, updating and combining dictionaries, using an IPL season example to illustrate multi-level data structures.

Learn how to create dictionaries from two lists using the zip method in Python, wrapping it with dict to convert keys and values into a usable mapping.

Learn how tuples work in Python: treat as immutable pairs with fixed length, accessible by indexes, where you cannot modify or add elements once created.

Explore sets in Python: learn about unordered collections with no duplicates, how to add, discard, or remove elements, and how update merges with another set while noting they're not indexable.

Learn to implement multiple conditional branches in Python using if, elif, and else, comparing variables like x and y and handling complex decision logic for practical machine learning tasks.

Learn how while loops use a condition to repeat code and avoid infinite loops, with break and increments. Explore printing 1 to 10 and even numbers using modulo.

Explore how the Python range() function generates number sequences, from zero-based starts to one-based starts, with practical examples like 1 to 10 and 5 to 10.

Learn how for loops execute while a condition holds, iterate over lists to print elements in order, and perform simple operations on strings and numbers during iteration.

Learn to avoid using reserved keywords as Python variable names, and see how importing the keyboard library and using try and lambda help manage these terms.

Learn Python's built-in functions for strings and numbers, including upper, lower, capitalize, title, and sum. Discover how to sort with sort and sorted, using reverse and key with lambda.

Define and call functions in Python and C with clear parameters. Use blocks and conditional statements inside functions, and utilize returns to compute and pass values.

Explore python lambda functions as anonymous, single-line expressions that replace named functions, showing max value logic for two and three variables, and converting multi-line code into concise one-liners.

Explore file I/O essentials in practical machine learning with Python by opening, reading, writing, and closing text files, using read, readline, seek, and the with statement.

Learn what numpy is and why it matters in machine learning with Python. Explore how zip and list comprehension compare to traditional methods in nanoseconds, highlighting numpy as the umpire.

Explore how to create one-dimensional and multi-dimensional numpy arrays, inspect shapes, dimensions, and data types, and manipulate elements to understand array metadata.

Master broadcasting in Python for practical machine learning by aligning array shapes with new axes to enable compatible operations.

Explore numpy built-in functions for array operations, including max, min, argmax, nonzero elements, boolean indexing, random number generation, and floor and ceiling computations.

Explore data types such as integers (signed and unsigned), 8/16/32/64-bit, boolean, string, and unicode. Learn to use a types property to inspect and specify exact bit widths to save space.

Learn how to convert data types in Python, changing strings to numbers and ensuring compatibility between types, using type casting to prepare data for machine learning workflows.

Learn how to perform matrix multiplication with matching dimensions to produce an output, compare matrix-based calculation to direct numeric multiplication, and note efficiency considerations.

Learn how to reshape numpy arrays from one-dimensional to four by three, transpose to swap rows and columns, and revert to the original shape using practical Python techniques.

Learn to slice NumPy arrays with Python syntax to select specific rows and columns, modify elements, and understand one dimensional and two dimensional slicing patterns for practical data manipulation.

Apply boolean indexing on a numpy array to select values using conditions like <=. Explore one- and multi-dimensional indexing and filter out Microsoft, generating windows from five to eight.

Learn to filter data with numpy by applying conditions such as less than six, greater than six, and use where to fetch corresponding x and y values from arrays.

Master mean and median calculations in Python, perform row-wise and column-wise means, and verify the overall data totals for practical machine learning.

Discover how to sort a numpy array and quickly retrieve its minimum and maximum values using index-based methods and np.max and np.min.

Discover how to stack and split arrays with vstack, hstack, and split methods, reshape data across dimensions, and produce multivariable outputs for practical python machine learning workflows.

Demonstrate the difference between view and copy in Python: shared memory causes changes to propagate between x and y, while copy creates independent memory; verify identity with is.

Learn how pandas series work within a data frame, including default and custom indexes. Convert dictionaries to series and fetch values to understand data alignment in Python.

Learn how to create a pandas data frame, generate random numbers for one- and multi-dimensional arrays, and build data frames from dictionaries with specified rows, columns, and index.

Learn how to view dataframe metadata in pandas by using head and tail, listing columns and indices, and checking shape and dtypes, and describe statistics plus null checks with isnull.

Learn how to rename columns and indices in an existing data frame, verify changes by listing columns, and apply these techniques for practical machine learning tasks in Python.

Explore transposing a dataframe in Pandas by using the .T attribute to swap rows and columns, with a hands-on example of three columns and six rows.

Learn how to slice a dataframe by selecting specific rows and columns using index positions and column names, including combining row and column selections for precise data retrieval.

Learn how to use boolean indexing on a dataframe to filter rows with conditions like less than or equal to 5, combine multiple conditions, and select specific values using isin.

Explore practical strategies for handling missing values in a data frame, including dropping rows or columns, filling with zeros or missing, forward and backward fill, mean, and interpolation.

Learn how to replace values in a dataframe using the Pandas replace method, including in-place updates, replacements via lists and dictionaries, and ensuring complete mappings.

Extract string data from a pandas DataFrame, filter to captains, and create new columns for movie name and year using string methods and regular expressions.

discover how to set and unset index, reindex data with new values, and apply index operations on single or multiple columns with practical examples.

Explore built-in and customized functions in pandas to compute column and row statistics, apply row/column operations, and create new columns, including a celsius-to-fahrenheit transformation using apply and lambda.

Learn how the value_counts() method computes frequencies in Python, sorts them in descending order, and shows the frequency of items such as Greek letters.

Group data by animals in Excel and Python, applying aggregations like max, min, mean, and sum. Create new columns with group by results and use aggregate or transform.

Practice concatenating data frames in pandas and appending data points across rows or across columns, using axis to switch between vertical and horizontal joins, illustrated with Excel examples.

Learn how to merge data frames using simple, left, right, and full outer joins on common columns, with Python examples that merge mobility and revenue data across continents.

Learn to reshape data frames with stack() and unstack(), handling index and column alignment, and resetting the index to return to the original form.

Explore pivoting data frames to create pivot tables of revenue by region (Asia, America, Europe) and compare formats for easy cross-checking and data processing.

Learn to reshape and clean data in pandas using melt, set index, rename, and stack techniques to prepare movie revenue datasets for practical machine learning with Python.

Convert categorical variables to dummy variables to feed machine learning algorithms, using get_dummies to create numeric features from string data.

Explore how the crosstab() method builds a frequency table from a data frame of movies and countries, turning countries into columns and movies into rows.

Clean messy data in Python by applying string methods such as upper, extract, replace, and split, using regular expressions to remove special characters and extract movie names.

Explore regular expressions for cleaning data with pandas: extract numbers, names, and alphabets; use quantifiers like * and +, square brackets for multi-column results, and ignore special characters.

Apply the contains() method to search strings, extract content, and obtain boolean results, using keywords, literals, and expressions to locate data in Python.

Apply the startswith method to filter a list of countries by starting letters, with examples showing China starts with C.

Apply multiple string methods at once in Python to remove leading and trailing spaces with strip and manipulate characters for clean text.

Explore how to manipulate column names in pandas by removing spaces, replacing spaces with underscores, and converting names to uppercase or title case, with practical examples.

Learn to filter a Python data frame by selecting columns whose names start with a given keyword, enabling you to extract targeted comments and values.

Learn to read csv files with pandas read_csv, handle missing headers, assign column names, skip comments, and set the correct index to construct a clean data frame.

Discover how to read tab-delimited files by treating tab as the separator, recognizing headers and columns, and using the current approach instead of deprecated methods to define five columns.

Learn how to read fixed-width files with pandas, using read_fwf and various parameters to handle separators and file formats for reliable data loading.

Learn to create JSON files and read JSON data from a URL, local file, or variable, including how to access and record outputs for machine learning workflows.

Learn to scrape HTML data from the web using pandas, with practical examples from Wikipedia and hands-on practice reading lists, elements, and data streams.

Learn to convert xml data into usable Python structures with dedicated functions and libraries, save results to variables, and work with dictionaries during web scraping.

Master web scraping by making API calls, sending GET and POST requests, and converting responses to JSON for data extraction.

Export a data frame to a csv file in python using a library, then save the results and verify the output by opening the file.

Learn how to load encoded data files in Python, handle encoding issues like utf-8 and Unicode decode errors, and use trial-and-error with encoding parameters to ensure proper data loading.

Identify error tokens, trailing commas, and other bad data when loading a file with Python and pandas, and either correct the file or ignore bad lines for large datasets.

Learn to select columns by data type using include and exclude patterns, filtering bools, dates, and other types to tailor your dataset for machine learning workflows.

Convert non timestamp data to timestamp data in pandas using to_datetime on object date columns, enabling year, month, day operations and creating combined datetime columns.

Learn how to handle invalid data when converting to datetime in pandas, comparing errors ignore and errors coerce options, turning invalid entries into nat or filtering them.

Convert unix time to a standard timestamp by interpreting seconds since 1970-01-01 and applying accurate date conversion, even with unusual or invalid looking date columns.

learn how to use pandas to create data frames with a datetime index by converting date lists to a datetime index and using set index.

Extract the current date and time in Python using a straightforward method, and apply it to current data.

Explore how to generate date ranges with date_range() and bdate_range() in pandas, including start and end dates, frequency, and business days, with examples across days, months, minutes, and seconds.

Learn how to slice a pandas series using a time index to retrieve specific values or ranges. Adjust the data length to explore index-based access and date-based slicing.

Explore slicing pandas dataframes with date or time indices, selecting date ranges, reversing order, and retrieving alternating values, while comparing slicing with truncate for efficient data access.

Explore datetime handling in pandas by converting values to datetime, creating date columns and series, and applying boolean indexing with multiple conditions, including between two and three.

Explore how to convert datetime data using the strftime method in pandas, formatting timestamps with codes like %Y, %m, and %d to produce custom string representations.

This lecture contrasts period range and date range in pandas and datetime, showing how to generate a period index or a datetime index with a chosen frequency.

Learn to use pandas period objects and frequency to increment years and days, distinguish period from timestamp, and explore time-based durations in notebooks.

Explore resampling in pandas to generate data with specific frequencies (e.g., every five minutes), work with index and columns, and apply forward fill to handle missing ranges.

Learn to handle time zones in pandas by creating time-indexed data, checking for timezone info, and using localize and utc conversion on series and data frames.

Learn to plot data on a single axis by importing pandas and generating x-axis values automatically, using sample data such as 100, 200, 300, 400, 500.

Learn how to plot a graph with two axes by mapping data to the x axis and data squares to the y axis, illustrating relationships between data and squares.

Learn to change line style and color of a plot, adjust markers like circles, triangles, and squares, and explore default colors and available styles using quick interactions and documentation.

Learn to limit x and y axis values in plots by setting x limits, such as 0 to 15 or 0 to 40, or providing both axes in a list.

Learn how to adjust the line width of a plot in Python, experimenting with values like 3.5 and 4.0 to improve readability and visualization clarity.

Learn to plot multiple lines in the same graph using Python, including creating data, computing squares and cubes, and customizing colors, while combining two plots into one.

Learn how to add a title and x and y labels to a plot using Python plotting tools, and customize axis labels and the plot title for clear visualization.

Learn how to enable or disable grid lines in a plot, including turning them on or off and understanding the default setting.

Learn to add annotations to a plot with pandas, using text at specific x and y coordinates and proper axis references.

Explore generating numeric sequences for plots and customizing x and y ticks, including interval, start and end, decimals, and rotating tick labels to improve readability.

Learn to customize a plot by managing spines: hide top and right spines, and move left and bottom spines to the center using the set position method.

Learn how to add a legend to a matplotlib plot using the label argument to distinguish lines for squares and cubes, enabling clear, labeled data visualization.

Explore building multiple plots with the subplot method, arranging graphs in two by two grids, sharing x or y axes, and mixing line, scatter, and axis configurations for synchronized visuals.

Learn to draw a line plot in Python using pandas to load a CSV data set, group by species, and customize color, axis rotation, and title.

Learn to draw bar graphs in Python, customizing with titles, color maps, and axis orientation, using data with species and petal columns, and exploring fixed sizes and stacking options.

Import pandas, create a scatter plot by supplying numeric x and y columns; avoid non-numeric data to prevent errors and learn basic scatter plot requirements.

Explore drawing area plots in Python with matplotlib, transforming a line plot into an area plot using transparency and a title, based on random data generation.

Explore box plots to summarize data distributions using count, mean, median, min, max, and percentiles, identify outliers, and compare groups by species.

Learn to visualize data distributions by creating histograms from a data frame column using pandas, adjust bins from 10 to 20, and filter values using boolean indexing.

Learn to create and customize pie charts in Python using pandas and plotting libraries, including labels, autopct percentages, explode effects, subplots, start angle, shadow, and figure sizing.

Learn how to create a seaborn count plot from the iris data using pandas to compute value counts, convert to a data frame, and rotate x-axis labels with subplots.

Explore how to create and customize a box plot using seaborn and pandas, plotting movie revenue by movie name with adjustable axes, rotation, and figure size.

Learn to create violin plots with Seaborn from a Pandas data frame of movie revenue, comparing distributions to box plots, and customize with subplots and axis labeling.

Explore how to create and customize a swarm plot with pandas, observe how it displays individual data points and counts, and compare its differences to other plotting approaches.

Learn to overlay plots in python by combining seaborn plots, such as violin and swarm plots, on the iris dataset with color customization for clearer data insights.

Learn to create univariate distribution plots with Seaborn, comparing histograms and kernel density estimates, and customize with bins, rug plots, and multi-plot layouts for numeric columns.

Explore facet grid visualizations with seaborn to create distribution plots of iris species, customizing by species with separate plots or shared axes using map, row, and column parameters.

Learn how to create linear relationship plots with Seaborn's lmplot and regplot, compare single plots with grid and facet configurations, and display multiple species using column or row layouts.

Adjust plot size and shape using the size and aspect parameters, set width and height, explore square and rectangular layouts, and manage subplots to control the plotting area.

Learn how to build and interpret pair plots in Python, combining scatter plots and histograms for multiple variable combinations to quickly overview a data frame.

Explore joint plot in python to visualize x vs y with combined scatter plots and histograms, using kde and distribution plots to reveal relationships and separate distributions.

Explore computing correlations among variables and visualizing them with a Seaborn heat map using Iris dataset. See how positive and negative relationships appear and how color maps reflect their strength.

Explore how data fall into qualitative and quantitative types, with discrete and continuous subtypes, and distinguish nominal, ordinal, and binary data for practical machine learning applications.

Explore the difference between population and sample by taking a random sample of 15 rows from a 150-row dataset, and see how the sample relates to the full population.

Explore sampling methods in practical machine learning, distinguishing biased vs unbiased approaches and applying examples of convenience, response, simple random, stratified random, and multi-stage sampling.

Explore the two branches of statistics—descriptive statistics describe existing data and inferential statistics predict or draw inferences about unseen data, using measures like mean, minimum, maximum, median, and standard deviation.

Analyze how distribution reveals performance patterns by comparing means and scores, using plots to distinguish uniform, normal, and skewed distributions for player selection.

Explore variance and standard deviation as spread measures, compare mean, median, and mode for central tendency, and note when to use iqr for skewed data.

Understand z-scores that standardize scores by mean subtraction and division by standard deviation to compare exams, with 68%, 95%, and 99.7% rules.

Explore how to measure and interpret correlation in wine data using pandas and seaborn, examining positive, negative, and neutral correlations and the role of the correlation coefficient.

Explore how machine learning models derive formulas from data, compare linear, exponential, and logistic approaches, and evaluate supervised versus unsupervised algorithms for practical prediction.

Explore fundamental probability through coin tosses and ball draws, computing heads, tails, and green balls, then apply conditional probability and Bayes' theorem to compute probabilities.

Distinguish labeled from unlabeled data, explain independent versus dependent variables and features, and introduce univariate and multivariate, as well as supervised versus unsupervised learning.

Explore the main machine learning types—supervised and unsupervised—by contrasting regression and classification, and outlining algorithms like linear, polynomial, decision tree, random forest, logistic regression, knn, svm, clustering, and anomaly detection.

The lecture explains supervised learning with labeled and unlabeled data, partitions 70–80 for training and 20–30 for evaluation, fits a linear regression model, and uses accuracy to predict.

Use count vectorizer to convert categorical data to features and create columns. Train and evaluate model on labeled data, then apply it to unlabeled data with contract tracing in Python.

Learn the difference between fit and fit_transform in regression modeling. See how fit computes model parameters (m and c) and how fit_transform fits and transforms the data.

Remove numerical and special characters from text using pandas, load data, lower-case the review column, apply a regular expression to replace non alphabetic characters with spaces, and split into words.

Learn to clean text data by removing HTML tags with Python, using a lambda function and list comprehension, and applying regex to keep only letters and reduce noise.

Learn how to remove stop words from text using Python, build a stop words list, and apply it to an entire column with list comprehension and a lambda function.

Apply a stemmer from a Python library to reduce related words to common roots, using list comprehension to stem plurals and deduplicate features.

Split labeled data into 70/30 or 80/20 training and testing sets in Python, train a model on the larger portion, and evaluate accuracy on the smaller test portion.

Compare regression accuracy using MAE, MSE, RMSE, and the variance (R^2) score, explain how to compute true versus predicted errors, and choose efficient algorithms with dataset splits.

Practically demonstrates simple linear regression using years of experience as the independent variable, splitting data into train and test, fitting the model, and evaluating with MAE, RMSE, and R-squared.

illustrates the difference between simple and multiple linear regression, separates eleven independent variables from the wine quality dependent variable, and demonstrates dataset preparation and a model score.

apply polynomial regression to predict kilowatts per month from home size using polynomial features and degree four, achieving a high score around 0.95.

Implement classification algorithms, including support vector classifier, multinomial naive Bayes, random forest, and decision tree, by preprocessing text data, converting to numerical features, performing a train-test split, and evaluating accuracy.

Learn how to implement clustering algorithms in Python, including data cleaning, creating dummy variables, applying k-means and hierarchical clustering, labeling clusters, and evaluating linkage methods.