beginner to advanced - how to become a data scientist

Learn to create a series from lists or dictionaries, define and inspect the index, distinguish attributes from methods, and perform basic statistics, sorting, membership tests, and broadcasting in pandas.

Learn how to load, inspect, and transform data with pandas for exploratory analysis. Read a csv with pandas, view head and tail, check shape and dtypes, and describe statistics.

Explore how to inspect a Pandas data frame, view its index, columns, and values, and use info to understand shape, data types, and memory usage.

Learn to prune a data frame by selecting only the needed columns, inspect with head, and build simple or multi indices to speed up slicing on large data sets.

Discover practical pandas techniques for data scientists: slice data frames, select single or multiple columns, handle spaces in column names, and create new columns with assignment or insert.

Explore broadcasting operations on data frames, performing column-wise and row-wise arithmetic using add, subtract, multiply, and divide, and computing sums across selected columns.

Explore data analysis with pandas by loading a baseball dataset, inspecting with head and info, checking shape and missing data, and counting team occurrences with value_counts.

Discover how to handle missing values by dropping rows with any missing data using dropna, with how='all' or default, and subset options, before imputation in the next video.

Explore handling missing values by deciding when to drop rows or columns, selecting relevant columns, and using fill methods or backfill to preserve data for analysis.

Use S-type methods to convert data types in a data frame, turning strings into numbers and numbers into strings, while handling formats and in-place updates.

Sort a data frame by a column or index using sort values and sort index. Learn to set ascending or descending orders and handle missing values.

Explore data slicing with boolean masks to filter a Titanic data frame by sex and embarked, using pandas to inspect filtered data and the survival column as a binary target.

Explore advanced data frame filtering by applying greater than, less than or equal, and multi-criteria logic to Titanic passenger data, including sex and class, using pipe operators for or conditions.

Investigate missing values at the start of data exploration, use booleans to filter for missing age or cabin data, and compare filtering versus dropping to assess data quality.

Explore a practical machine learning case study on the Titanic dataset, applying data cleaning, missing value handling, and a decision tree classifier to train and predict survival.

Learn to convert string columns to date-time objects in R, using parse_dates and to_datetime, then filter data with between masks for start times and year ranges.

Learn how to handle duplicates in data frames using drop duplicates with subset and keep options, explore unique values with unique and nunique methods, and clean your dataset efficiently.

Learn to change a dataframe's index by using a name column as the index with set_index, and reset the index or use the index column parameter to configure the index.

Learn to slice a data frame by label with loc, set the index to a name column, and retrieve rows or columns by labels, ranges, or lists, with membership checks.

Explore slicing and accessing data with loc, iloc, and ix, including index and column selection, masks, and 0-based counting. Set or reset the index and rename columns.

Learn to clean and explore data in pandas by dropping columns with drop, pop, or del, drawing random samples, and selecting the top or bottom records by a metric.

Learn to query pandas data using the query method, filtering by quoted strings and combining conditions with and or, including equality, inequality, and checks with names like Kadabra and Charizard.

Learn to work with dates using pandas and datetime, converting strings to date time objects, creating time stamps, and building date ranges and time indices for analysis.

Explore using Pandas date range creation and the dt accessor to test start or end of month, quarter, or year, and filter series with boolean masks.

Learn to work with timestamps and date time, extract year, month, day, and weekday, and create date ranges with start, end, and frequency using date offset.

learn to slice date ranges in pandas using start and end of month, quarter, and year, and filter series with boolean masks based on date states.

Learn to group data in pandas using group by, then aggregate with size, sum, mean, and more to analyze the IPL dataset by city or multiple columns.

Learn to work with a multiindex: set multi indices from name and type, sort and rename levels, swap levels, access level values, transpose, and melt the frame for visualization.

Learn how data science meets finance by importing stock data with a data reader, retrieving open, high, low, close, and volume, and slicing panels for single or multiple stocks.

learn to combine dataframes in pandas using append, concat, and merge; align data on index or date, resolve mismatches with suffixes, and merge frames on common columns.

Learn to manipulate strings in a data frame using string methods such as lower, upper, and title, and extract currency with regular expressions.

In this panda tips and tricks session, learn to reduce memory usage by dropping unused columns and converting repetitive columns to category, with caveats for unique fields.

Identify and fill missing values in the type 2 column, then use the lock method to update safely and avoid the setting with copy warning.

Learn to use map, apply, and apply map on a data frame to transform booleans with a dictionary, extract first letters, and compute maximums with built-in functions.

Learn how tensors extend beyond vectors and matrices to multidimensional arrays, and master shape, dimensions, dtype, and item size using NumPy arrays.

Explore numpy basics by creating arrays, performing element-wise arithmetic (add, subtract, multiply, divide, exponent), and using range and the type function with sin and cos.

Master NumPy indexing, slicing, and element modification for one-dimensional and two-dimensional arrays; create from lists, inspect shape and size, and reshape for machine learning tasks.

Learn how numpy arrays behave: slicing creates a reference to memory, so changes in a slice alter array; create arrays from lists and use ones, zeros, identity, empty, and fill.

Explore NumPy basics: sum across axes, mean max min functions, 1d and 2d arrays, broadcasting, argmax and argmin for indexes, and flattening and transposing data for machine learning.

Explore the basics of matplotlib for Python: create figures, plot and scatter data, customize axis limits, labels, and titles, and manage subplots for multi-panel visuals.

Explore advanced matplotlib techniques to visualize financial data, including using styles like 538 and ggplot, plotting close prices, horizontal bars, fill_between, and color maps.

Master advanced Matplotlib for 3d plotting with mpl_toolkits.mplot3d, including creating 3d axes, plotting lines with diverse styles and colors, and using legends.

Master Seaborn as a high-level, customizable visualization tool for data scientists; learn its aesthetic plots, data frame compatibility, and its value for data exploration, statistical visualization, and analysis.

Master seaborn by loading a pokemon dataset into a data frame, then plotting in a notebook with inline visuals built from seaborn.

Master seaborn basics in a Jupyter notebook by loading a dataset, creating a data frame, and plotting within the notebook using seaborn for clear, in-notebook visualizations.

Master Seaborn lm plots to visualize scatterplots, inspect correlation between attack and defense, and color points by stage while toggling regression lines and confidence intervals.

Master seaborn by customizing axes, setting zero-based limits, and plotting colored third-dimension visuals while disabling the regression line for clearer data storytelling.

Learn to create seaborn box plots to visualize data distributions, identify the median and quartiles, and handle outliers by filtering non-numeric columns like total, stage, and legendary.

Learn how to create seaborn violin plots to visualize distribution by category, customize style and colors with palettes, adjust figure size, and interpret the distribution by type and attack.

Master seaborn swarm plots with a single line of code (sns.swarmplot) to display each point, stack similar values, and tailor figure size, background, grid, and color palettes for clear distributions.

Combine violin and swarm plots in seaborn by adjusting figure size, mapping x and y to data frame columns, turning off inner bars, applying color with alpha, and setting title.

Master seaborn by transforming pokemon attributes into a dataset with pandas melt, keeping name, type 1, and type 2 while uniting attack, defense, speed, and hp under a statistics column.

Master seaborn swarm plots by preparing data with pandas melt, creating a value and an attribute column, and visualizing multiple types by color with legend placement and axis customization.

Explore seaborn heat maps to visualize a correlation matrix from a data frame, showing how HP correlates with other attributes and how to customize color palettes for clearer insights.

Master seaborn by visualizing distributions of categorical variables with count plots and counterplots, rotate x tick labels by 45 degrees, and adjust figure size for clearer, readable charts.

Learn to create three-panel Seaborn visualizations by coloring by state and faceting by stage with swarm plots, and apply consistent x-axis label rotation across all panels.

Learn to visualize relationships with Seaborn by using density plots and joint plots to explore two variables, such as attack and defense, including overlaying plots and interpreting correlation and p-values.

Complete the course and gain a deep understanding of data exploration and preparation, pi's importance for data science, and visualising findings with seaborn, ready for machine learning in Python.

Explore transfer learning using a pre-built neural network to predict ice cream. Apply preprocessing and decode predictions to identify likely flavors from images.