The repository contains all proposed solutions for the laboratory exercises completed during the a.y. 2024-2025 of the Data Visualization Lab course.

• Data manipulation

• Pandas and Numpy

• Inkscape

• Matplotlib

• Seaborn

• Ggplot

• Dimensionality reduction

• Geospatial mapping

• Bokeh

Here is a more detailed list of the topics covered in each lab:

• Lab 1: Cleaning of data using pandas

  • melt
  • pivot

• Lab 2: Data exploration/modification using pandas

  • lotr
  • concat
  • combining datasets with join and its variants

• Lab 3: Inkscape

• Lab 4: Plotting with matplotlib

  • matlab like interface (state-based)
  • object oriented interface
  • figure and subplots
  • axes vs axis
  • format string for data points and lines
  • shared x/y axis
  • figure size, aspect ratio, DPI and saving
  • labels and legends
  • formatting: LaTeX, rcParams, plot styles

• Lab 5: Plotting with matplotlib #2

  • colors, line widths, line types (similar to format string)
  • line and marker styles
  • axis appearance: range, ticks, tick labels
  • custom legend
  • texts, arrows and annotations
  • subplots layout: subplot2grid, GridSpec
  • histograms, boxplots, bar plots, error bars and continuous errors

• Lab 6: Plotting with pandas & seaborn

  • indexing and selection (column and row)
  • condition-based filtering
  • plotting using 'kind' and transpose method
  • area chart, histograms, bar charts, pie charts, boxplots, scatter plots
  • 'standard' plots, density plots, joint plots (with hue variable), pair plots

• Lab 7: Plotting with seaborn and further case studies

  • 'standard' plots, strip/swarm plot, catplot, linear regresssion, heatmaps, cluster maps
  • subplots with seaborn
  • faceted plots (show specific subgroups of the whole data set)
  • grid layouts: pairplot (auto), PairGrid (map plot functions to the empty grids)
  • case studies: encircling, diverging bars (and texts), diverging dot plot, cleveland dot plot, ridgeline plot, waffle chart, treemap, cluster plot, andrews plot, parallel coordinates

• Lab 8: Plotting with ggplot (plotnine)

  • data, aes(thetics), geometry (geom_* or stat_*), axes and themes
  • 'standard' plots, aes color
  • trend lines: geom_smooth and possible methods
  • layering
  • axis limits: xlim/ylim (deletion of points) vs. coord_cartesian (zoom region)

• Lab 9: Plotting with ggplot (plotnine) #2

  • scaling and scaling color
  • 'standard' plots
  • statistical transformations
  • position arguments (e.g stacked/dodge bars)
  • coordi_flip for horizontal drawing
  • themes (axis, ticks, legend, background), labels and element_*()
  • saving plots
  • distributions plotting (density, reference lines)
  • faceting

• Lab 10a: Dimensionality reduction

  • PCA
  • t-SNE
  • UMAP

• Lab 10b: Geospatial mapping

  • Geopandas
  • Geoplot
  • Plotly
  • Folium

• Lab 11: Bokeh: quick summary

During the lessons, the notebook files contained some exercises in order to test the understanding. About Inkscape, as there is no dedicated notebook file, a folder has been created inkscape where it is possible to see the original plots before_lab03.png and the final result lab03.png.

Each notebook includes exercises to test understanding of the topics covered. For Inkscape (Lab 3), since there is no dedicated notebook, a folder named inkscape contains the original plots showed during the lecture. In particular, in the output-svg folder it is possible to visualize the first version (before_lab03.png) and the final version (lab03.png) which has been generated for the assignment.

The folder Practice_exam contains an example of exam, which is given as a notebook file. All instructions are already written on it.

The folder Exam contains the exam of 12/06, which is given as a notebook file. All instructions are already written on it.

Given the amount of modules needed for each notebook, a requirements.txt has been generated so that it can be used to install directly all the modules needed, simply type in your terminal:

pip install -r requirements.txt

Dorijan Di Zepp dorijan.dizepp@studenti.unitn.it