Python — Data visualization (Matplotlib, Seaborn, Plotly)

Las preguntas del cuestionario

1. 1. Difference between Figure and Axes in Matplotlib?
   1. None
   2. Figure is the container; Axes is an individual plotting area with its own scales/ticks
   3. Axes contains the Figure
   4. Figure is deprecated
2. 2. Recommended way to create two side-by-side subplots?
   1. fig, (ax1, ax2) = plt.subplots(1, 2)
   2. plt.figure(); plt.figure()
   3. plt.subplot() twice with no args
   4. plt.draw(2)
3. 3. Best chart to compare the distribution of a continuous variable across groups?
   1. Pie chart
   2. Boxplot or violin plot
   3. Stacked bar chart
   4. 3D histogram
4. 4. Which chart is generally discouraged for comparing proportions across >4–5 categories?
   1. Bar chart
   2. Lollipop chart
   3. Pie chart
   4. Treemap
5. 5. Best chart to visualize the correlation between two continuous variables?
   1. Scatter plot
   2. Bar chart
   3. Pie chart
   4. Frequency heatmap
6. 6. How do you display a Matplotlib figure in a regular script?
   1. plt.show()
   2. plt.render()
   3. plt.display()
   4. It auto-displays
7. 7. What does this do?

   import matplotlib.pyplot as plt
   fig, ax = plt.subplots()
   ax.plot([1, 2, 3], [1, 4, 9])
   ax.set_yscale('log')

   1. Logarithmic Y axis
   2. Error
   3. Logarithmic X axis
   4. Bar chart
8. 8. Why save as .svg or .pdf instead of .png?
   1. Smaller files
   2. Vector format: stays crisp at any zoom (print, high-DPI screens)
   3. No benefit
   4. PNG lacks color support
9. 9. What does plt.tight_layout() do?
   1. Toggles fullscreen
   2. Auto-adjusts margins to prevent label/title overlap
   3. Removes axes
   4. Forces a square figure
10. 10. Which library produces interactive, web-ready charts (D3/WebGL-based)?
    1. Matplotlib
    2. Seaborn
    3. Plotly
    4. Pandas plotting
11. 11. Main advantage of Seaborn over raw Matplotlib?
    1. Faster
    2. High-level API for statistical plots + native Pandas integration + polished defaults
    3. Replaces Matplotlib
    4. Generates interactive PDFs
12. 12. Which palette family suits unordered categories?
    1. Sequential (viridis, plasma)
    2. Diverging (coolwarm)
    3. Qualitative (tab10, Set2)
    4. Cyclic
13. 13. Which palette fits continuous data with a meaningful midpoint (e.g., anomalies ±)?
    1. Sequential (viridis)
    2. Diverging (RdBu, coolwarm)
    3. Qualitative (tab20)
    4. Grayscale
14. 14. Why prefer viridis over jet?
    1. Personal taste
    2. viridis is perceptually uniform and colorblind-friendly; jet introduces artifacts and false banding
    3. jet is more colorful
    4. viridis does not exist
15. 15. How do you plot a Pandas Series directly?
    1. s.plot()
    2. plot(s)
    3. pd.show(s)
    4. Not possible without Matplotlib
16. 16. What does this produce?

    import seaborn as sns
    import pandas as pd
    df = pd.DataFrame({'x': [1,2,3,4], 'y': [2,4,1,8], 'g': ['a','a','b','b']})
    sns.scatterplot(data=df, x='x', y='y', hue='g')

    1. Scatter plot colored by group g
    2. Histogram
    3. Error: hue must be numeric
    4. Boxplot
17. 17. Why are titles, axis labels (with units), and legends essential?
    1. Decoration only
    2. They provide context, units, and meaning of colors/markers — required for readability
    3. They slow rendering
    4. Paper-only
18. 18. How do you avoid chartjunk?
    1. Add more colors and 3D effects
    2. Maximize the data-to-ink ratio: remove frames, backgrounds, and unnecessary effects (Tufte)
    3. Use background images
    4. Animate every element
19. 19. What's the issue with a bar chart whose Y axis doesn't start at 0?
    1. Nothing
    2. It visually exaggerates differences and can mislead readers — avoid for bars (acceptable for line charts with clear annotation)
    3. Improves precision
    4. Breaks PNG export
20. 20. Which Python library implements the grammar of graphics (ggplot-style)?
    1. Plotnine
    2. Seaborn only
    3. Matplotlib low-level
    4. Pandas