feat(seaborn): implement contour-filled

[github-actions]

Implementation: contour-filled - python/seaborn

Implements the python/seaborn version of contour-filled.

File: plots/contour-filled/implementations/python/seaborn.py

Parent Issue: #2500

---

🤖 impl-generate workflow

[claude]

AI Review - Attempt 1/3

Image Description

Light render (plot-light.png): A filled contour plot with warm off-white background (#FAF8F1) displaying a temperature field across longitude and latitude. The plot uses the viridis colormap with smooth color gradients ranging from deep purple (low temperature ~1°C) to bright yellow (high temperature ~33°C). Three distinct temperature peaks appear as concentric color bands with yellow cores fading to cooler blues. The title 'contour-filled · seaborn · anyplot.ai' is clearly visible in dark text at the top. Axis labels 'Longitude (°E)' and 'Latitude (°N)' are descriptive with units. A colorbar labeled 'Temperature (°C)' is positioned on the right. Subtle contour lines overlay the filled regions at alpha=0.3. All text elements (title, axis labels, tick labels) are rendered in appropriate dark ink tones and are completely readable against the light background. Grid is removed for a clean appearance.

Dark render (plot-dark.png): The same plot rendered on a warm near-black background (#1A1A17). All text elements have been theme-adapted to light colors for readability: title and axis labels are in bright off-white (#F0EFE8), tick labels in softer light ink (#B8B7B0). The data colors in the filled contours and colorbar are identical to the light render—the viridis colormap is perceptually uniform and theme-independent. The colorbar numbers and label are clearly visible in light text. All chrome elements (background, text, axis spines, grid) have adapted to the dark theme, while the data representation remains consistent. No dark-on-dark legibility issues detected.

Score: 88/100

Category	Score	Max
Visual Quality	30	30
Design Excellence	12	20
Spec Compliance	15	15
Data Quality	15	15
Code Quality	10	10
Library Mastery	6	10
Total	88	100

Visual Quality (30/30)

• VQ-01: Text Legibility (8/8) - All font sizes explicitly set; title 24pt, axis labels 20pt, tick labels 16pt; perfectly readable in both themes
• VQ-02: No Overlap (6/6) - No overlapping text; x-axis labels, y-axis labels, and colorbar are all spaced appropriately
• VQ-03: Element Visibility (6/6) - Filled contours and contour lines are sharp and distinct; three temperature peaks are clearly visible with excellent visual separation
• VQ-04: Color Accessibility (2/2) - Viridis is perceptually uniform and colorblind-safe; good contrast between temperature regions
• VQ-05: Layout & Canvas (4/4) - Plot fills 50-80% of canvas with balanced margins; colorbar integrated well; no wasted space
• VQ-06: Axis Labels & Title (2/2) - Descriptive labels with units: 'Longitude (°E)', 'Latitude (°N)', colorbar shows 'Temperature (°C)'
• VQ-07: Palette Compliance (2/2) - Viridis colormap correct for continuous data; backgrounds are warm off-white (#FAF8F1) light and warm near-black (#1A1A17) dark; text colors theme-correct in both renders

Design Excellence (12/20)

• DE-01: Aesthetic Sophistication (5/8) - Well-executed styling with theme tokens properly applied; clean professional appearance but relies on library defaults rather than custom design
• DE-02: Visual Refinement (4/6) - Grid removed for cleanliness; contour lines subtly styled with alpha=0.3; colorbar well-integrated; minimal embellishment follows minimalism principle
• DE-03: Data Storytelling (3/6) - Realistic temperature field data is inherently meaningful; color progression intuitively maps temperature but lacks explicit visual hierarchy or emphasis beyond natural data structure

Spec Compliance (15/15)

• SC-01: Plot Type (5/5) - Correct filled contour implementation with contour line overlay
• SC-02: Required Features (4/4) - All spec features present: filled contours, colorbar, contour lines, sequential colormap, grid resolution (80×80 points)
• SC-03: Data Mapping (3/3) - X: Longitude, Y: Latitude, Z: Temperature; axes show full data range
• SC-04: Title & Legend (3/3) - Title format 'contour-filled · seaborn · anyplot.ai' correct; colorbar labeled with units

Data Quality (15/15)

• DQ-01: Feature Coverage (6/6) - Shows all aspects: three distinct hot peaks, cool regions, smooth gradients, full colormap range
• DQ-02: Realistic Context (5/5) - Temperature field across geographic region is realistic, neutral, and comprehensible scenario
• DQ-03: Appropriate Scale (4/4) - Temperature range (1–33°C) is plausible; Gaussian decay is geographically sensible

Code Quality (10/10)

• CQ-01: KISS Structure (3/3) - Imports → Data → Plot → Save; no unnecessary functions or classes
• CQ-02: Reproducibility (2/2) - Uses np.random.seed(42); deterministic meshgrid
• CQ-03: Clean Imports (2/2) - Only used imports: os, matplotlib, numpy, seaborn
• CQ-04: Code Elegance (2/2) - Clean, Pythonic code; appropriate complexity; no fake UI
• CQ-05: Output & API (1/1) - Saves as 'plot-{THEME}.png' with correct DPI=300

Library Mastery (6/10)

• LM-01: Idiomatic Usage (4/5) - Proper use of sns.set_theme(), sns.color_palette(), fig.colorbar(); theme tokens applied idiomatically; good matplotlib integration
• LM-02: Distinctive Features (2/5) - Limited distinctive seaborn features; core visualization uses matplotlib (ax.contourf, ax.contour); appropriate for this plot type since seaborn lacks built-in contour

Score Caps Applied

• None - no caps triggered

Strengths

• Perfect Visual Quality: All text explicitly sized and perfectly readable in both light and dark themes; no overlap or visibility issues; excellent theme adaptation
• Complete Spec Implementation: All required features present including filled contours, contour line overlay, colorbar, and sensible grid resolution
• Realistic & Neutral Data: Temperature field is a comprehensible, real-world scenario (weather/climate data) with plausible values and geographic structure
• Clean Code: KISS structure, reproducible with seed, no unnecessary complexity; proper theme-token implementation throughout
• Correct Palettes & Colors: Viridis colormap is perceptually uniform and CVD-safe; theme backgrounds and text colors are correctly implemented in both renders

Weaknesses

• Design Excellence Below Exceptional: DE-01 relies on library defaults rather than custom design choices; lacks distinctive visual sophistication or intentional design hierarchy
• Limited Data Storytelling: Plot displays data effectively but doesn't emphasize insights through visual techniques (color contrast, size variation, focal points); storytelling derives from data structure rather than design
• Seaborn Features Underutilized: Core plotting is handled by matplotlib (ax.contourf); seaborn role is primarily theme management rather than distinctive library features

Issues Found

1. DE-01 MODERATE: Generic styling with library defaults (viridis is standard, no custom palette or design intent)
   • Rationale: Implementation is correct but not exceptional; would benefit from intentional design choices to elevate aesthetic sophistication
2. DE-03 LOW: Data is well-chosen but visual hierarchy relies on natural data variation rather than design emphasis
   • Rationale: No explicit visual hierarchy, emphasis, or visual guide to direct the viewer's attention to key insights

AI Feedback for Next Attempt

If this were to undergo revision: Consider adding visual hierarchy through strategic data emphasis (e.g., highlight the hottest region with annotation or contrast), refine typography with intentional weight variation, or adjust colorbar styling to add visual sophistication. However, current implementation is strong and passes quality thresholds.

Verdict: APPROVED

---

Threshold Met: Score 88 ≥ 80 (Review 2 threshold)
This implementation is approved for merge. It demonstrates strong technical execution, complete spec compliance, and proper theme adaptation across both light and dark renders.

[github-actions]

🔧 Repair Attempt 1/4

Applied fixes based on AI review feedback.

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

---

🤖 impl-repair

[claude]

AI Review - Attempt 1/3

Image Description

Light render (plot-light.png): The plot shows a filled contour of a geographic temperature field on a warm off-white background (#FAF8F1). Three distinct features are visible: two warm peaks (red/dark-red) labeled "Warm Peak A ≈32°C" and "Warm Peak B ≈30°C", and one cold trough (bright blue) labeled "Cold Trough ≈4°C". The colormap appears to be a diverging red-blue scheme (not the viridis specified in the code), with mid-range temperatures rendering as light salmon/pink. A colorbar on the right is labeled "Temperature (°C)" with readable tick values. The title "contour-filled · seaborn · anyplot.ai" and axis labels "Longitude (°E)" / "Latitude (°N)" are all clearly legible against the light background. All text is readable. Legibility verdict: PASS.

Dark render (plot-dark.png): The outer figure background is near-black, consistent with #1A1A17. The title, axis labels, and tick labels are all rendered in light/muted tones (correctly adapted). Annotation boxes have dark fills with light text. Data colors (warm red peaks, cold blue trough) are identical to the light render — only chrome flips. One note: the mid-range contour colors (light salmon/pink for ~10°C regions) create a mostly-light center area within the dark-themed figure, which is the expected data representation but creates a slight visual contrast with the dark frame. No dark-on-dark failures found. Legibility verdict: PASS.

Critical discrepancy noted: The rendered images show (1) three annotation boxes with leader lines ("Warm Peak A", "Warm Peak B", "Cold Trough") and (2) a diverging red-blue colormap, neither of which exist in the current seaborn.py code. The images appear to have been generated from an earlier version of the implementation. The current code uses sns.color_palette("viridis", as_cmap=True) with no annotations.

Score: 82/100

Category	Score	Max
Visual Quality	26	30
Design Excellence	12	20
Spec Compliance	15	15
Data Quality	15	15
Code Quality	9	10
Library Mastery	5	10
Total	82	100

Visual Quality (26/30)

• VQ-01: Text Legibility (7/8) — All font sizes explicitly set (title 24pt, labels 20pt, ticks 16pt); both renders readable. Minor: title lacks bold weight.
• VQ-02: No Overlap (6/6) — Clean layout, no collisions.
• VQ-03: Element Visibility (5/6) — Contourf fills are prominent; contour overlay lines (alpha=0.3, linewidth=0.5) are very subtle but acceptable.
• VQ-04: Color Accessibility (2/2) — Continuous colormap with colorbar reference; no red-green sole signal.
• VQ-05: Layout & Canvas (3/4) — Good proportions with set_aspect('equal') for geographic data; colorbar well-positioned.
• VQ-06: Axis Labels & Title (2/2) — All labels include units; title matches required format exactly.
• VQ-07: Palette Compliance (1/2) — Plot backgrounds are theme-correct per code. However, rendered images show a diverging red-blue colormap (likely coolwarm/RdBu) rather than the viridis specified in code or the recommended viridis/cividis/BrBG from the style guide. Code-image discrepancy (see Image Description).

Design Excellence (12/20)

• DE-01: Aesthetic Sophistication (5/8) — Geographic temperature field with distinct warm/cold regions shows above-default design thinking. Annotation boxes (in rendered images) add informational value. Not yet publication-ready: spines retained, no explicit typographic hierarchy.
• DE-02: Visual Refinement (3/6) — Grid explicitly removed; contour overlay lines are tastefully subtle; tight_layout applied. Missing: spine removal (sns.despine()), explicit whitespace optimisation.
• DE-03: Data Storytelling (4/6) — Two warm peaks + cold trough create clear focal points; geographic context is compelling. Annotation boxes (visible in rendered images) guide the viewer to key features. Without annotations (current code), storytelling relies entirely on the colormap contrast.

Spec Compliance (15/15)

• SC-01: Plot Type (5/5) — Correct filled contour plot.
• SC-02: Required Features (4/4) — 15 levels (within spec's 10–20 range), colorbar, contour line overlay, smooth gradient.
• SC-03: Data Mapping (3/3) — Longitude × Latitude grid (80×80, within spec's 30–100), Temperature as Z-value; correct axis assignment.
• SC-04: Title & Legend (3/3) — Title is exactly "contour-filled · seaborn · anyplot.ai"; colorbar labeled "Temperature (°C)" with units.

Data Quality (15/15)

• DQ-01: Feature Coverage (6/6) — Demonstrates filled contour, overlay isolines, colorbar reference, and distinct warm/cold regions.
• DQ-02: Realistic Context (5/5) — Geographic temperature field is a classic, neutral, real-world application of filled contour plots.
• DQ-03: Appropriate Scale (4/4) — Longitude −180→180, latitude −90→90, temperature 1–31°C — all physically plausible.

Code Quality (9/10)

• CQ-01: KISS Structure (3/3) — Linear flat script; no functions or classes.
• CQ-02: Reproducibility (2/2) — np.random.seed(42).
• CQ-03: Clean Imports (2/2) — os, matplotlib.pyplot, numpy, seaborn all used.
• CQ-04: Code Elegance (1/2) — Two issues: (a) sns.color_palette("viridis", as_cmap=True) is an unusual roundabout to get a viridis colormap — use cmap='viridis' directly; (b) colorbar label coloring is done twice redundantly (cbar.ax.tick_params(colors=INK_SOFT) and then the for label loop).
• CQ-05: Output & API (1/1) — plot-{THEME}.png correct; current seaborn 0.13 API.

Library Mastery (5/10)

• LM-01: Idiomatic Usage (3/5) — sns.set_theme() with full RC token dict is exemplary seaborn theming. The actual contourf rendering falls back to matplotlib (seaborn has no native contourf), which is the correct approach for this plot type.
• LM-02: Distinctive Features (2/5) — Seaborn's theme system is a genuinely distinctive contribution. sns.color_palette(as_cmap=True) shows library-specific knowledge. However, no seaborn statistical/plotting functions are used (inherent to this plot type). Consider sns.despine() as an additional seaborn-idiomatic touch.

Score Caps Applied

• None

Strengths

• Perfect spec compliance: all required features present (levels, colorbar, overlay contour lines, correct grid resolution)
• Excellent data quality: geographic temperature field is realistic, neutral, and showcases the full range of filled contour capabilities
• Theme-adaptive chrome is properly wired via sns.set_theme() RC params — both renders pass theme readability check
• KISS code structure with reproducible seed; all font sizes explicitly set
• Correct output filename convention

Weaknesses

• Code-image discrepancy: Current seaborn.py has no annotations and specifies viridis, but rendered images show a red-blue diverging colormap and three annotation boxes — images appear stale from a previous code version
• VQ-07 colormap mismatch: Rendered colormap is not among approved options (viridis, cividis, BrBG); use cmap='viridis' directly in ax.contourf()
• Design refinement gap: No spine removal; spines should be removed with sns.despine(); title should be bold weight
• Redundant colorbar styling: both cbar.ax.tick_params(colors=INK_SOFT) and the for label loop do the same thing — remove the loop
• Library mastery ceiling: Seaborn's statistical API is not used; add sns.despine() and consider whether any seaborn-specific features (e.g. kdeplot for a synthetic 2D density scenario) could strengthen the seaborn identity

Issues Found

1. VQ-07 LOW: Rendered colormap does not match style guide recommendations.
   • Fix: Replace palette = sns.color_palette("viridis", as_cmap=True) with direct cmap='viridis' argument in ax.contourf(..., cmap='viridis'). Remove the palette variable entirely.
2. DE-01/DE-02 GAP: Missing spine removal and title bold weight reduce visual polish.
   • Fix: Add sns.despine(ax=ax) after plotting; add fontweight='bold' to ax.set_title().
3. CQ-04 REDUNDANCY: Colorbar label color set twice.
   • Fix: Remove the for label in cbar.ax.get_yticklabels() loop; cbar.ax.tick_params(colors=INK_SOFT) is sufficient.
4. DE-03 STORYTELLING: Current code has no annotations despite the rendered images showing them — add ax.annotate() calls for the warm peaks and cold trough to guide the viewer.

AI Feedback for Next Attempt

Fix colormap: replace sns.color_palette('viridis', as_cmap=True) with cmap='viridis' directly in ax.contourf(). Add sns.despine(ax=ax) for visual refinement and fontweight='bold' to the title. Remove the redundant colorbar label loop. Add 2-3 ax.annotate() calls labeling the temperature extremes (warm peaks and cold trough) to improve data storytelling. These targeted fixes should close the gap to 90+.

Verdict: REJECTED

[github-actions]

🔧 Repair Attempt 1/4

Applied fixes based on AI review feedback.

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

---

🤖 impl-repair

[claude]

AI Review - Attempt 2/3

Image Description

Light render (plot-light.png): The plot renders on a warm off-white background (#FAF8F1) showing a global temperature field as a filled contour plot. The viridis colormap runs from deep purple (cold, ~1°C) through teal and green to bright yellow (warm, ~32°C). Two warm peaks are visible — one in the south-central Atlantic (~100°W, 40°S) and one in Asia-Pacific (~100°E, 20°N) — and one cold trough at northern mid-latitudes (~30°E, 70°N). Subtle contour isolines are overlaid at 0.5 linewidth and 0.3 alpha. Two annotation boxes with arrows label the "Cold Trough ≈4°C" and "Warm Peak ≈32°C". The title "contour-filled · seaborn · anyplot.ai" is dark and clearly readable at 24pt. Axis labels "Longitude (°E)" and "Latitude (°N)" are dark at 20pt; tick labels at 16pt. Colorbar "Temperature (°C)" is properly labeled with readable tick values. All text is readable against the light background.

Dark render (plot-dark.png): The same plot on a warm near-black (#1A1A17) background. The viridis data colors are identical to the light render — the same purple cold region and yellow warm peaks — confirming Okabe-Ito continuity (positions 1–7 unchanged). Title, axis labels, and tick labels are all rendered in light ink and are clearly readable against the dark background. Colorbar label and tick values remain legible. Annotation boxes adapt to the dark theme, with readable text. No dark-on-dark failures observed; no near-black text on near-black background. Both backgrounds are correctly warm (not pure black / pure white).

Both paragraphs are required. A review that only describes one render is invalid.

Score: 88/100

Category	Score	Max
Visual Quality	28	30
Design Excellence	14	20
Spec Compliance	15	15
Data Quality	15	15
Code Quality	10	10
Library Mastery	6	10
Total	88	100

Visual Quality (28/30)

• VQ-01: Text Legibility (7/8) — Title 24pt, axis labels 20pt, ticks 16pt all correctly sized; readable in both themes
• VQ-02: No Overlap (5/6) — Annotation boxes positioned without major collisions; minor tightness near upper plot edge
• VQ-03: Element Visibility (6/6) — Filled contours, isoline overlay, and colorbar all clearly visible in both themes
• VQ-04: Color Accessibility (2/2) — viridis is perceptually uniform and CVD-safe; no red-green sole signal
• VQ-05: Layout & Canvas (4/4) — 16:9 canvas, equal aspect ratio appropriate for geographic coordinates
• VQ-06: Axis Labels & Title (2/2) — Longitude (°E), Latitude (°N), colorbar Temperature (°C) — all descriptive with units
• VQ-07: Palette Compliance (2/2) — viridis correct for continuous data; #FAF8F1 light and #1A1A17 dark backgrounds both correct

Design Excellence (14/20)

• DE-01: Aesthetic Sophistication (5/8) — Professional geographic temperature field with intentional design; annotations add hierarchy
• DE-02: Visual Refinement (4/6) — Grid removed, equal aspect, subtle isoline overlay; all 4 spines retained and annotation boxes use default matplotlib styling rather than theme-adaptive ELEVATED_BG tokens
• DE-03: Data Storytelling (5/6) — Annotations with arrows pointing to Cold Trough and Warm Peak create clear focal points and guide the viewer effectively

Spec Compliance (15/15)

• SC-01: Plot Type (5/5) — Correct filled contour plot via contourf
• SC-02: Required Features (4/4) — Filled contours, isoline overlay, colorbar, 80×80 grid — all spec requirements met
• SC-03: Data Mapping (3/3) — Longitude/Latitude/Temperature correctly mapped; full axis ranges displayed
• SC-04: Title & Legend (3/3) — Title "contour-filled · seaborn · anyplot.ai" exact format; colorbar serves as legend with "Temperature (°C)"

Data Quality (15/15)

• DQ-01: Feature Coverage (6/6) — Shows filled bands, isolines, colorbar, multiple peaks and troughs — full feature showcase
• DQ-02: Realistic Context (5/5) — Global temperature field is realistic, scientifically plausible, and neutral
• DQ-03: Appropriate Scale (4/4) — 80×80 grid (within 30–100 spec range); temperatures 1–32°C realistic; full-globe coordinates

Code Quality (10/10)

• CQ-01: KISS Structure (3/3) — No functions or classes; clean linear script
• CQ-02: Reproducibility (2/2) — np.random.seed(42)
• CQ-03: Clean Imports (2/2) — os, matplotlib.pyplot, numpy, seaborn — all used
• CQ-04: Code Elegance (2/2) — Appropriate complexity, no fake UI
• CQ-05: Output & API (1/1) — Saves as plot-{THEME}.png with facecolor=PAGE_BG

Library Mastery (6/10)

• LM-01: Idiomatic Usage (4/5) — sns.set_theme() with rc dict is idiomatic seaborn theming; core contourf is matplotlib (unavoidable for this plot type)
• LM-02: Distinctive Features (2/5) — sns.color_palette("viridis", as_cmap=True) is seaborn-specific but the plot is fundamentally matplotlib; limited seaborn distinctiveness given plot type constraints

Score Caps Applied

• None

Strengths

• Perfect spec compliance — filled contour with overlaid isolines (15 levels), colorbar with label, viridis colormap, 80×80 grid resolution
• Excellent data storytelling via annotations with arrows pointing to Cold Trough (≈4°C) and Warm Peak (≈32°C)
• Both light (#FAF8F1) and dark (#1A1A17) themes render cleanly with all text readable and correct backgrounds
• Appropriate viridis colormap for continuous temperature field — perceptually uniform, CVD-safe, style-guide compliant
• Clean code with seed for reproducibility, proper theme-adaptive tokens, correct save pattern

Weaknesses

• All four spines retained (full box frame) — removing top+right spines (L-shaped) would be more refined per style guide
• Annotation boxes use default matplotlib styling (neutral grey fill, default border) rather than theme-adaptive ELEVATED_BG tokens (#FFFDF6 light / #242420 dark)
• seaborn library mastery inherently limited for this plot type — sns.set_theme and sns.color_palette are the main seaborn contributions; core plot is matplotlib

Issues Found

1. DE-02 MEDIUM: Annotation boxes use matplotlib default styling instead of theme tokens
   • Fix: Style annotation boxes with bbox=dict(boxstyle='round,pad=0.4', facecolor=ELEVATED_BG, edgecolor=INK_SOFT, alpha=0.9) for both annotations
2. LM-02 LOW: Limited seaborn-specific API usage
   • Fix: Already maximizing seaborn capabilities for this plot type; sns.set_theme and color_palette are the correct seaborn patterns here

AI Feedback for Next Attempt

Strong implementation with excellent spec compliance and data storytelling. The annotations are a meaningful addition. To push higher: (1) style annotation boxes using theme tokens — bbox=dict(boxstyle='round,pad=0.4', facecolor=ELEVATED_BG, edgecolor=INK_SOFT, alpha=0.9) — so they blend with the elevated surface rather than using matplotlib defaults; (2) optionally remove top+right spines if geographic frame isn't strictly needed, or at minimum keep them but verify spine colors use INK_SOFT; (3) seaborn library mastery is already at ceiling for this plot type given contourf is a matplotlib-only API.

Verdict: APPROVED