feat(altair): implement scatter-annotated

[github-actions]

Implementation: scatter-annotated - altair

Implements the altair version of scatter-annotated.

File: plots/scatter-annotated/implementations/altair.py

---

🤖 impl-generate workflow

[claude]

AI Review - Attempt 1/3

Image Description

The plot displays a scatter plot with 10 tech company data points showing Revenue (Billions USD) on the x-axis (ranging from 0-640) and Profit Margin (%) on the y-axis (ranging from 0-60). Each point is rendered as a blue circle (#306998) with moderate opacity. Company names are displayed as bold dark text labels positioned to the upper-right of each point. Companies shown include NVIDIA (high profit margin ~55%, low revenue ~61B), Apple (high revenue ~385B, moderate margin ~25%), Microsoft, Amazon (highest revenue ~574B, lowest margin ~6%), Google, Meta, Adobe, Oracle, Tesla, and Intel. The title "scatter-annotated · altair · pyplots.ai" appears centered at the top. Grid lines are subtle with low opacity. The overall layout is clean and balanced.

Quality Score: 91/100

Criteria Checklist

Visual Quality (36/40 pts)

• VQ-01: Text Legibility (10/10) - Title is large (28pt), axis labels are 22pt, tick labels are 18pt, all clearly readable at full size
• VQ-02: No Overlap (8/8) - No overlapping text; labels are well-positioned with dx=12, dy=-8 offset
• VQ-03: Element Visibility (8/8) - Marker size=250 with opacity=0.7 is appropriate for 10 data points
• VQ-04: Color Accessibility (5/5) - Single blue color (#306998) with good contrast, no colorblind issues
• VQ-05: Layout Balance (3/5) - Good overall balance, but data clusters in left-center leaving right side sparse (this is data-driven, not a layout issue per se)
• VQ-06: Axis Labels (2/2) - Descriptive with units: "Revenue (Billions USD)" and "Profit Margin (%)"
• VQ-07: Grid & Legend (0/2) - Grid is subtle (0.3 opacity), but no legend is present (though not strictly needed for this single-series plot)

Spec Compliance (24/25 pts)

• SC-01: Plot Type (8/8) - Correct scatter plot with text annotations
• SC-02: Data Mapping (5/5) - X=revenue, Y=profit_margin correctly assigned
• SC-03: Required Features (4/5) - Has text labels near points; however, spec mentions "subtle connecting lines or arrows from labels to points when offset" which is missing
• SC-04: Data Range (3/3) - Axes show all data with appropriate padding (0-620 for x, 0-60 for y)
• SC-05: Legend Accuracy (2/2) - No legend needed for single series, N/A
• SC-06: Title Format (2/2) - Correct format: "scatter-annotated · altair · pyplots.ai"

Data Quality (20/20 pts)

• DQ-01: Feature Coverage (8/8) - Shows variation: high margin/low revenue (NVIDIA), low margin/high revenue (Amazon), and middle performers
• DQ-02: Realistic Context (7/7) - Real tech companies with realistic financial metrics; neutral business topic
• DQ-03: Appropriate Scale (5/5) - Revenue and profit margin values are realistic for these companies

Code Quality (9/10 pts)

• CQ-01: KISS Structure (3/3) - Clean structure: imports → data → points layer → labels layer → combine → save
• CQ-02: Reproducibility (3/3) - Uses np.random.seed(42), though data is actually deterministic arrays
• CQ-03: Clean Imports (2/2) - Only altair, numpy, pandas - all used
• CQ-04: No Deprecated API (0/1) - Using current Altair API
• CQ-05: Output Correct (1/1) - Saves as plot.png and plot.html

Library Features (2/5 pts)

• LF-01: Uses distinctive library features (2/5) - Uses Altair's layered grammar (points + labels), tooltip encoding, and declarative style. However, could leverage more Altair-specific features like interactive selections, conditional encodings, or text adjustments

Strengths

• Excellent text legibility with appropriate font sizes for high-resolution output
• Clean declarative layer composition using Altair's grammar of graphics
• Realistic, neutral dataset with recognizable tech companies and accurate financials
• Smart label positioning with consistent dx/dy offsets to avoid point overlap
• Proper use of tooltips for interactivity in HTML output

Weaknesses

• Missing connector lines/arrows from labels to points as mentioned in specification notes
• Grid/Legend scoring reduced since no legend is present (though arguably not needed for single-series)
• Could leverage more Altair-specific features like conditional formatting or selections

Verdict: APPROVED