Data analysis and visualization in Python with Pandas

Explore how pandas computes statistical descriptions and simple aggregations. Use sum, axis options, and describe to obtain mean, std, min, max, and percentiles (20%, 50%, 70%) while handling NaN values.

Describe how pandas handles non-numerical data, showing count, unique, top, and freq, and summarize methods like min, max, argmin, argmax, idxmin, idxmax, sum, mean, median, var, std, diff, and pct_change.

Explore using the isin method to filter unique values and build a boolean mask, then apply pandas value_counts across Q1–Q3 with zeros for non-numeric values.

Learn to manipulate missing data in pandas by filtering with dropna, using a series containing none values, and observing how dropna drops missing entries while preserving valid data.

Learn to filter missing data in pandas by using isnull and dropna to drop rows or columns with NaN, including how='all' to drop only all-NaN rows or columns.

Learn how to fill missing data in pandas using fillna with scalar values, per-column fills via a dictionary, and mean (average) imputation for a series in a data frame.

Explore hierarchical indexing in pandas, using multi-level indexes and unstack to transform a multi-index series into a data frame, with practical examples of indexing, selecting, and unstacking.

Explore hierarchical indexing in pandas by building a two-level index with two-level columns, renaming levels, swapping them, and summing by level and by columns.

Learn to work with dataframe columns in pandas by setting and resetting indices, creating multi-index dataframes, and understanding how columns become index in dataframes.

Create a pandas data frame from cattle, chicken, and sheep, indexed by country; use lambda to find min and max, then compute sum and average; sort by index and sheep.

Explore the clustering exercise and build a pandas data frame from cattle, chicken, and sheep, perform filtering with loc, compute min, max, sums, means, and handle missing values.

Learn how to read and write data from files in Python using Pandas, including read_csv and read_table, handling headers, custom column names, and index columns.

Learn to read and write data from files with pandas, including multi-index creation using a list of index columns, handling missing values with na_values, and skipping rows with read_csv.

Learn to perform partial reading of large text files with pandas by reading in chunks using read_csv and chunk size, then compute value counts by key.

Learn how to write data to text formats using pandas, exporting with to_csv and reading with read_csv, and control headers and delimiters for csv files.

Read excel files into pandas data frames using xlrd and openpyxl, load a file, parse sheet one, and produce a data frame with columns A, B, C, D, and message.

Learn json data concepts and handling in python by using the json library to load and dump data, and pandas to build dataframes from json with selected fields.

Read 1880, 1881, 1882 into data frames df_1880, df_1881, df_1882. Count 1880 births by gender and Alice; Grace across frames; set 1180–1200 in df_1880 to ABC; list df_1882 last names.

Learn to load three year data frames with pandas read_csv, filter by gender, sum totals, and count names like Grace and Alice.

Explore data visualization in Python using matplotlib and seaborn, learn to plot lines with various colors, markers, and styles, and apply auto scale and tight layout for clear figures.

Create a scatter graph in Python by plotting x and y values with Matplotlib, then customize size, color, marker type, and transparency to illustrate data points.

Create and customize a bar graph in python using two one-dimensional arrays for x and y, including default and full-form bars, plus horizontal bars with colors and edge colors.

Explore creating a pie graph with matplotlib using a one-dimensional data array, customizing with labels, explode, colors from seaborn palettes, auto percent, and an optional shadow.

Learn to create a single figure with multiple subplots on a 2x2 grid, placing bar, pie, and scatter plots in separate panels, and apply seaborn color palettes with panel titles.

Position titles, ticks, and labels in a matplotlib plot using set_title, set_x_label, set_y_label, and set_x_ticks/set_y_ticks, with custom tick labels and rotation.

Learn how to position legends in matplotlib to identify multiple graphs, using legend and loc=best to automatically place the legend where space is greatest.

Explore line drawing in pandas using the series and data frame plot methods, with matplotlib and numpy that plot indexed values and base ten logs.

Explore bar drawing in pandas by plotting bar and horizontal bar charts with kind='bar', using series from linspace data and a figure with two graphs.

Create a scatter plot in python using pandas and matplotlib by loading macro data, selecting CPI, M1, tb rate, and unemployment, and plotting M1 differences against unemployment with labeled axes.

Read macro data and marker data csv files into data frames, create frames a, b, and c with year, real gdp, and real cons, and plot the graph for 1960–1963.

Read macro data from csv files with pandas to build data frames. Visualize real GDP and related metrics using line plots and horizontal bar charts.

Master data wrangling in Python with Pandas by reading, cleaning, transforming, and rearranging data, and learn how to combine datasets with merge and concat operations.

Merge data frames using pandas' merge method to link rows by keys. Learn inner, left, right, and outer joins and how to use on, left_on, and right_on for column names.

Learn to merge data frames in Python using left and outer joins on two keys, and resolve overlapping columns with suffixes.

Explore merging dataframes by index and by column in pandas, using left_on and right_index, including multi-index joins and the outer join to align all indexes.

Learn pandas concatenation with pd.concat to merge series and data frames along different axes, illustrating vertical and horizontal stacking and index/column alignment.

Explore how to concatenate data frames in pandas using pd.concat, control join behavior with axis and inner join, and build hierarchical indexes with keys.

Learn how hierarchical indexing enables data frame rearrangement in pandas, using stack and unstuck to rotate between rows and columns.

Learn to remove duplicate data in pandas by examining a dataframe, identifying duplicates with the duplicated method, and using drop_duplicates to keep non-duplicate rows.

Learn how to transform data in pandas by mapping values through a dictionary and using lambda functions, including lowercasing text and creating a new animal column.

Learn to replace values in pandas series using the replace method, including replacing with numpy NaN, handling multiple replacements, and using a dict to map old values to new ones.

Learn to rename pandas indexes using map and rename, converting indexes to upper or title case with str.upper and str.title, and adjust column labels using a deque structure.

Read Titanic data in pandas by loading train and test csv into dataframes, select pclass, sex, age, fare, embarked for C and D, perform left join to merge and concatenate.

Learn to analyze Titanic data with pandas by reading train and test csvs, selecting pclass, sex, age, fare, and embarked, and merging and concatenating frames using index operations.

Explore data aggregation with pandas groupby mechanics, splitting a dataframe by keys along rows or columns, applying functions, and combining results into grouped summaries.

Explore advanced group by mechanics in python with pandas, using two keys and arrays to build multi-index results, unstack for wide format, and compute mean values across UK and USA.

Learn group iteration in Python by grouping data by a key, yielding tuples with a group name and its data, and inspecting each group's key and data.

Explore column selection in pandas; compare selecting columns for aggregation with group by operations, including single and multi-column grouping, yielding series or data frames depending on brackets.

Group a dataframe by a dictionary mapping columns to red and blue, then sum values to compare red (a, b) and blue (c, d) groups.

Explore grouping with functions in Pandas, using len to group by index length and compute sums or minimum. Build multi index results; relate functions to lists, arrays, dictionaries, and series.

Explore data aggregation in pandas, transforming arrays into scalar values via group by and agg functions, including peak to peak (max minus min) and tips.csv data insights.

Group by columns in a pandas tips data frame to compute mean and other metrics (std, min, max, count) across sex and smoker, producing hierarchical, multi-index results.

Apply multiple functions to dataframe columns using a dictionary in pandas, aggregating typekit with min, max, mean, and std, and size with sum after grouping by sex and smoker.

Learn the split-apply-combine workflow in pandas, create a top function to select top tip pct values, and apply it with group by smoker and day.

Explore pivot tables in pandas to summarize and aggregate data by multiple keys, using pivot_table and group by, with margins and aggregate functions like count and sum.

Analyze Titanic data with pandas by loading Titanic.csv, selecting pclass, sex, age, fare, embarked, performing group by counts and aggregations, and identifying the oldest traveler without an index.

Analyze Titanic data with pandas by reading the csv, selecting five columns, grouping by sex and embarked, and applying min, max, sum, count, and size aggregations to reveal distributions.

Explore the fundamentals of time series data, its importance across finance, economics, ecology, and physics, and how pandas supports time stamps, separation, rearrangement, aggregation, and fixed frequencies.

Explore Python date and time data types using the datetime module. Import datetime, create current time, inspect year, month, and day, and compute differences with time delta between two datetimes.

Convert strings to date objects in Python using datetime, strftime, and strptime, with pandas to_datetime for date arrays and formatting codes like %Y-%m-%d.

Explore time series basics in pandas by creating datetime indices, building a series with timestamped values, inspecting the datetime index and dtype, and converting indices to timestamp objects.

Explore indexing and selection in Python with pandas, focusing on time series operations, date-like indices, slicing, and extracting year and month from data frames and series.

Explore time series with double indexes in pandas, learn how to handle duplicate date indices, check uniqueness, and select all entries matching a specific date like 2018-01-02.

Learn to convert irregular time series to a fixed daily frequency with pandas resample('D'), applying daily mean and handling missing days as NaN in the resulting series.

Learn to generate date time indexes with pandas date_range, using start, end, or periods, and apply daily (D) or last business day (BM) frequencies.

Explore how pandas uses base frequencies and multipliers, such as four hours, and calendar codes like D, H, T, S, M, BM to build date ranges.

Explore data replacement before and after using the shift method on time series and data frames, moving values forward and backward while preserving the index, with monthly and daily examples.

Explore periods and arithmetics in pandas, creating period objects for year-end monthly or quarterly intervals and using period range and period index. See additions or subtractions shift the period end.

Learn to convert period index objects to different frequencies in pandas, transitioning from annual December periods to monthly start and end indexes, and creating series with month-end and month-start indexes.

Convert timestamps to periods using the two period method on datetime indexes and series, illustrating how monthly frequency yields period indexes for January and February 2020.

Create a period index from year and quarter columns in pandas, with a q dec frequency, and assign it as the dataframe index.

Explore resampling and frequency conversion in pandas, focusing on downsampling high-frequency data to lower frequencies and upsampling to higher frequencies, with mean and sum aggregations.

Upsample by converting lower frequency data to higher frequency using resample and mean aggregation, turning weekly data into daily values and aligning date ranges.

Draw time series in pandas by loading stock data from csv and parsing dates; set a date index, select AAPL, MSFT, XOM, SP, fill missing values forward, and plot trends.

Depict a time series of Apple stock from 2005 to 2009 with simple, arithmetic, and exponential moving averages on a two-panel plot.