Data science is an interdisciplinary field that applies information from data across a wide range of application fields by using scientific methods, procedures, algorithms, and systems to infer knowledge and insights from noisy, structured, and unstructured data.
Data visualization libraries like matplotlib, Bokeh, bqplot, Plotnine, cufflinks, Altair, hvplot, Holoviews, seaborn and more.
Different Interactive charts Sunburst Charts, Sankey Diagrams (Alluvial), Candlestick Charts, Network Charts, Chord Diagram, Parallel Coordinates Plots, Radar Charts, Connection Map, Treemap, Choropleth Maps, Scatter & Bubble Maps.
Apart from this, you will find tutorials about time series data and its applications, creating dashboards, and other concepts.
For an in-depth understanding of the above concepts, check out the sections below.
Tutorial provides a brief guide to detect stationarity (absence of trend and seasonality) in time series data. After checking for stationarity, the tutorial explains various ways to remove trends and seasonality from time series to make them stationary.
A detailed guide on how to use Python library "cufflinks" to create interactive data visualizations/charts. Cufflinks is built on top of Plotly and let us create charts by calling 'iplot()' method on Pandas dataframe. The 'iplot()' method tries to mimic 'plot()' API (matplotlib) of pandas dataframe to generate charts but uses Plotly.
A detailed guide to creating Sankey Diagram (Alluvial Diagram) using Python data visualization libraries Plotly and Holoviews (Bokeh & Matplotlib). The charts are interactive and visualized in Jupyter Notebooks.
Tutorial explains how to use Python module "missingno" to analyze the distribution of missing data (NaNs/NULLs/None Values) in our datasets. It let us create various charts to visualize the spread of missing data from various angles which can help us make better decisions.
How to Create Basic Dashboard using Streamlit and Cufflinks (Plotly)?
Tutorial provides detailed guide on how we can use pivot() and pivot_table() function available from pandas to create pivot tables. The pivot_table() function also let us perform many simple stats on aggregate data.
Geoviews - Choropleth Maps using Bokeh and Matplotlib [Python]
Data Visualization is a field of graphical representation of information / data. It is one of the most efficient ways of communicating information with users as humans are quite good at capturing patterns in data.
Python has a bunch of libraries that can help us create data visualizations. Some of these libraries (matplotlib, seaborn, plotnine, etc) generate static charts whereas others (bokeh, plotly, bqplot, altair, holoviews, cufflinks, hvplot, etc) generate interactive charts. Majority of basic visualizations like bar charts, line charts, scatter plots, histograms, box plots, pie charts, etc are supported by all libraries. Many libraries also support advanced visualization, widgets, and dashboards.
Basic Data Visualizations like bar charts, line charts, scatter plots, histograms, box plots, pie charts, etc are quite good at representing information and exploring relationships between data variables.
But sometimes these visualizations are not enough and we need to analyze data from different perspectives. For this purpose, many advanced visualizations are developed over time like Sankey diagrams, candlestick charts, network charts, chord diagrams, sunburst charts, radar charts, parallel coordinates charts, etc. Python has many data visualization libraries that let us create such advanced data visualizations.
Dashboards are literally everywhere and everyone is using them. Dashboards are GUI with various visualizations and metrics that can be used to monitor key performance indicators. Dashboards have a very wide range of applications in all fields.
Python has a bunch of libraries (dash, panel, streamlit, bokeh, etc) that let us create dashboards using them. They let us include widgets and interactive data visualizations in dashboards.
Time series is a type of data where data points are recorded in time order or at specified time intervals. Many real-world datasets like stock prices, weather indicators, heights of ocean tides, retail sales, etc.
Time series analysis involves various tasks like resampling time series, trying moving window functions, forecasting, classification, etc.
Python has various libraries (pandas, statsmodels, etc.) that let us load and work with time series data efficiently. They even provide useful functionalities to work with time series data.
Maps are one of the best ways to display and analyze geospatial data. It helps us better see patterns and trends geographically. This can help us with better decision-making.
Many different types of maps have been developed over time to analyze data from different perspectives. Some common map visualization types are choropleth maps, scatter maps, bubble maps, connection maps, etc. Apart from these, we can also include pins on maps to identify locations.
Python has many different libraries (geopandas, folium, ipyleaflet, cartopy, geoviews, geoplot, bokeh, altair, plotly, hvplot, etc) that let us create static as well as interactive maps.
Exploratory data analysis (commonly referred to as EDA) is an initial analysis of data to look for various relationships, anomalies, missing values, distributions, basic statistics, etc. It helps us understand data better to make further decisions. Various stats are calculated and statistical visualizations are created during EDA to understand data.
Python provides many different tools / libraries (Sweetviz, missingno, seaborn, pandas, etc) for performing EDA. It's quite common to use more than one of these tools to perform EDA.