feat(matplotlib): implement scatter-matrix

[github-actions]

Implementation: scatter-matrix - matplotlib

Implements the matplotlib version of scatter-matrix.

File: plots/scatter-matrix/implementations/matplotlib.py

---

🤖 impl-generate workflow

[claude]

AI Review - Attempt 1/3

Image Description

The plot displays a 4×4 scatter plot matrix (pairplot) visualizing Iris-like flower measurement data across four variables: Sepal Length, Sepal Width, Petal Length, and Petal Width (all in cm). The diagonal cells contain stacked histograms showing the distribution of each variable for three species: Setosa (blue #306998), Versicolor (yellow #FFD43B), and Virginica (green #4CAF50). Off-diagonal cells display scatter plots showing pairwise relationships between variables, with points colored by species. A legend is positioned in the top-right subplot. The title "scatter-matrix · matplotlib · pyplots.ai" appears at the top. Grid lines are visible with subtle styling. Variable names with units appear along the left and bottom edges.

Quality Score: 85/100

Criteria Checklist

Visual Quality (33/40 pts)

• VQ-01: Text Legibility (7/10) - Text is readable but font sizes are smaller than recommended (title 20pt vs 24pt, labels 13pt vs 20pt, ticks 11pt vs 16pt)
• VQ-02: No Overlap (8/8) - No overlapping text elements
• VQ-03: Element Visibility (6/8) - Markers visible but s=40 is on the smaller side; works reasonably for 150 points but could be slightly larger
• VQ-04: Color Accessibility (5/5) - Blue, yellow, and green palette is colorblind-friendly with good differentiation
• VQ-05: Layout Balance (3/5) - Square 12×12 figure is appropriate for matrix, but some wasted space between subplots
• VQ-06: Axis Labels (2/2) - All labels include units "(cm)"
• VQ-07: Grid & Legend (2/2) - Grid subtle (alpha=0.3, dashed), legend well placed

Spec Compliance (25/25 pts)

• SC-01: Plot Type (8/8) - Correct scatter plot matrix/pairplot
• SC-02: Data Mapping (5/5) - Variables correctly mapped to rows/columns
• SC-03: Required Features (5/5) - Histograms on diagonal, scatter on off-diagonal, color encoding for categories, axis labels on edges
• SC-04: Data Range (3/3) - All data visible within axes
• SC-05: Legend Accuracy (2/2) - Species names match data correctly
• SC-06: Title Format (2/2) - Correct format "{spec-id} · {library} · pyplots.ai"

Data Quality (17/20 pts)

• DQ-01: Feature Coverage (7/8) - Shows correlations, clusters, and species separation well; Petal measurements clearly separate Setosa
• DQ-02: Realistic Context (7/7) - Iris dataset is the classic example for scatter matrices
• DQ-03: Appropriate Scale (3/5) - Values are realistic for flower measurements, though species parameters could show more overlap/variation

Code Quality (10/10 pts)

• CQ-01: KISS Structure (3/3) - Clean imports → data → plot → save structure
• CQ-02: Reproducibility (3/3) - np.random.seed(42) set
• CQ-03: Clean Imports (2/2) - Only matplotlib and numpy used
• CQ-04: No Deprecated API (1/1) - No deprecated functions
• CQ-05: Output Correct (1/1) - Saves as plot.png

Library Features (0/5 pts)

• LF-01: Uses distinctive library features (0/5) - Pure matplotlib implementation without using pandas.plotting.scatter_matrix or other advanced features

Strengths

• Excellent implementation of the scatter matrix concept with proper histograms on diagonal and scatter plots off-diagonal
• Good use of color encoding to distinguish three species with a colorblind-friendly palette
• Clean, well-organized code following KISS principles
• Realistic Iris-like data that clearly demonstrates the purpose of scatter matrices
• Proper axis labeling with units on edges only (not repeated in every subplot)
• Symmetric matrix as specified

Weaknesses

• Font sizes are smaller than matplotlib library guidelines recommend (title 20pt vs 24pt, labels 13pt vs 20pt, ticks 11pt vs 16pt)
• Marker size s=40 is on the smaller side for 150 data points; s=60-80 would improve visibility
• Does not use distinctive matplotlib features like pandas integration or advanced subplot formatting
• Figure size 12×12 produces approximately 3600×3600 at dpi=300, which is correct for square format, but the internal spacing could be tighter

Verdict: REJECTED

Score of 85/100 is below the 90 threshold. Main issues are undersized text/markers and lack of distinctive library features.

[github-actions]

🔧 Repair Attempt 1/3

Applied fixes based on AI review feedback.

Status: Repair completed, re-triggering review...

---

🤖 impl-repair

[claude]

AI Review - Attempt 2/3

Image Description

The plot displays a 4×4 scatter plot matrix using the classic Iris flower dataset structure with four measurements: Sepal Length (cm), Sepal Width (cm), Petal Length (cm), and Petal Width (cm). The diagonal cells contain overlapping histograms showing the distribution of each variable, with three species color-coded: Setosa (blue #306998), Versicolor (yellow #FFD43B), and Virginica (green #4CAF50). Off-diagonal cells show scatter plots for each variable pair, clearly revealing the characteristic clustering patterns (Setosa well-separated from the other two species, especially in petal measurements). The legend is positioned in the top-right subplot. Axis labels with units appear on the left and bottom edges. A subtle dashed grid (alpha=0.3) appears in each subplot. The title follows the required format at the top.

Quality Score: 91/100

Criteria Checklist

Visual Quality (36/40 pts)

• VQ-01: Text Legibility (10/10) - Title at 24pt, axis labels at 20pt, tick labels at 16pt - all clearly readable
• VQ-02: No Overlap (8/8) - No overlapping text elements; layout is clean
• VQ-03: Element Visibility (6/8) - Markers at s=70 are appropriate for 150 points; histograms have good alpha=0.7; could be slightly larger for this data density
• VQ-04: Color Accessibility (5/5) - Blue/yellow/green palette is colorblind-safe, good contrast
• VQ-05: Layout Balance (5/5) - Square format (12x12) with tight spacing works well for matrix; plot fills canvas appropriately
• VQ-06: Axis Labels (2/2) - All labels include units: "Sepal Length (cm)" etc.
• VQ-07: Grid & Legend (0/2) - Grid is appropriate (alpha=0.3), but legend placement in top-right subplot partially overlaps with scatter points in that cell

Spec Compliance (25/25 pts)

• SC-01: Plot Type (8/8) - Correct scatter plot matrix with diagonal histograms
• SC-02: Data Mapping (5/5) - Pairwise relationships correctly mapped
• SC-03: Required Features (5/5) - All spec features present: diagonal distributions, off-diagonal scatters, color encoding for groups, variable labels on edges
• SC-04: Data Range (3/3) - All data visible within axis ranges
• SC-05: Legend Accuracy (2/2) - Species names correctly labeled with matching colors
• SC-06: Title Format (2/2) - "scatter-matrix · matplotlib · pyplots.ai" follows required format

Data Quality (20/20 pts)

• DQ-01: Feature Coverage (8/8) - Shows correlations, clusters (Setosa separation), different distribution shapes, varying relationships
• DQ-02: Realistic Context (7/7) - Iris-like flower measurements are a classic, realistic example
• DQ-03: Appropriate Scale (5/5) - Values realistic for flower measurements (sepal 4-8cm, petal 0-7cm, etc.)

Code Quality (10/10 pts)

• CQ-01: KISS Structure (3/3) - Linear flow: imports → data → plot → save (no functions/classes)
• CQ-02: Reproducibility (3/3) - np.random.seed(42) set
• CQ-03: Clean Imports (2/2) - Only matplotlib, numpy, GridSpec - all used
• CQ-04: No Deprecated API (1/1) - Uses current matplotlib API
• CQ-05: Output Correct (1/1) - Saves as plot.png

Library Features (0/5 pts)

• LF-01: Uses distinctive library features (0/5) - Uses basic GridSpec and manual loop; could use plt.subplots() more efficiently but doesn't leverage any advanced matplotlib features

Strengths

• Excellent implementation of the scatter plot matrix with clear visual separation of species clusters
• Proper use of GridSpec for tight subplot spacing creating a cohesive matrix appearance
• Appropriate use of transparency (alpha=0.6 for scatter, 0.7 for histograms) to handle overlapping points
• Clean axis label placement with labels only on edges (left and bottom)
• Well-chosen colorblind-safe palette with good contrast
• Realistic Iris-like data with biologically plausible parameters

Weaknesses

• Legend placement in top-right subplot overlaps with scatter data in that cell; consider placing legend outside the matrix or in a dedicated area
• Scatter marker size (s=70) is adequate but could be slightly larger for better visibility given the data density
• Could leverage more distinctive matplotlib features like shared axes linking or colorbar for additional encoding

Verdict: APPROVED