feat(altair): implement scatter-lag

plots/scatter-lag/implementations/altair.py

@@ -0,0 +1,112 @@
+""" pyplots.ai
+scatter-lag: Lag Plot for Time Series Autocorrelation Diagnosis
+Library: altair 6.0.0 | Python 3.14.3
+Quality: 90/100 | Created: 2026-04-12
+"""
+
+import altair as alt
+import numpy as np
+import pandas as pd
+
+
+# Data - synthetic AR(1) process with moderate autocorrelation
+np.random.seed(42)
+n_points = 500
+lag = 1
+phi = 0.85
+noise = np.random.normal(0, 1, n_points)
+values = np.zeros(n_points)
+values[0] = noise[0]
+for i in range(1, n_points):
+    values[i] = phi * values[i - 1] + noise[i]
+
+y_t = values[:-lag]
+y_t_lag = values[lag:]
+r_value = np.corrcoef(y_t, y_t_lag)[0, 1]
+
+df = pd.DataFrame({"y_t": y_t, "y_t_lag": y_t_lag, "time_index": np.arange(n_points - lag)})
+
+# Reference line (y = x diagonal)
+margin = 0.5
+axis_min = min(df["y_t"].min(), df["y_t_lag"].min()) - margin
+axis_max = max(df["y_t"].max(), df["y_t_lag"].max()) + margin
+ref_df = pd.DataFrame({"x": [axis_min, axis_max], "y": [axis_min, axis_max]})
+
+# Annotation for correlation coefficient
+annot_df = pd.DataFrame({"x": [axis_max - 0.3], "y": [axis_min + 0.5], "label": [f"r = {r_value:.3f}"]})
+
+# Reference line
+reference_line = (
+    alt.Chart(ref_df).mark_line(strokeDash=[8, 6], strokeWidth=1.5, color="#aaaaaa").encode(x="x:Q", y="y:Q")
+)
+
+# Scatter points with reduced size/opacity to prevent overplotting
+points = (
+    alt.Chart(df)
+    .mark_point(size=45, filled=True, strokeWidth=0.5, stroke="white", opacity=0.45)
+    .encode(
+        x=alt.X("y_t:Q", title="y(t)", scale=alt.Scale(domain=[axis_min, axis_max]), axis=alt.Axis(tickCount=10)),
+        y=alt.Y(
+            "y_t_lag:Q", title="y(t + 1)", scale=alt.Scale(domain=[axis_min, axis_max]), axis=alt.Axis(tickCount=10)
+        ),
+        color=alt.Color(
+            "time_index:Q",
+            scale=alt.Scale(scheme="viridis"),
+            legend=alt.Legend(
+                title="Time Index",
+                titleFontSize=16,
+                labelFontSize=16,
+                gradientLength=280,
+                gradientThickness=14,
+                orient="right",
+                offset=10,
+            ),
+        ),
+        tooltip=[
+            alt.Tooltip("y_t:Q", title="y(t)", format=".2f"),
+            alt.Tooltip("y_t_lag:Q", title="y(t+1)", format=".2f"),
+            alt.Tooltip("time_index:Q", title="Time Index"),
+        ],
+    )
+)
+
+# Correlation annotation
+annotation = (
+    alt.Chart(annot_df)
+    .mark_text(align="right", baseline="bottom", fontSize=20, fontWeight="bold", color="#333333")
+    .encode(x="x:Q", y="y:Q", text="label:N")
+)
+
+chart = (
+    (reference_line + points + annotation)
+    .properties(
+        width=1600,
+        height=900,
+        title=alt.Title(
+            "scatter-lag · altair · pyplots.ai",
+            fontSize=28,
+            subtitle=f"AR(1) process (φ = {phi}) | lag = {lag}",
+            subtitleFontSize=18,
+            subtitleColor="#666666",
+        ),
+    )
+    .configure_axis(
+        labelFontSize=18,
+        titleFontSize=22,
+        grid=True,
+        gridOpacity=0.15,
+        gridWidth=0.5,
+        domainWidth=0,
+        tickSize=6,
+        tickWidth=0.8,
+        tickColor="#999999",
+        labelColor="#444444",
+        titleColor="#333333",
+    )
+    .configure_view(strokeWidth=0)
+    .configure_title(anchor="start", offset=10)
+)
+
+# Save
+chart.save("plot.png", scale_factor=3.0)
+chart.save("plot.html")

plots/scatter-lag/metadata/altair.yaml

@@ -0,0 +1,238 @@
+library: altair
+specification_id: scatter-lag
+created: '2026-04-12T18:11:36Z'
+updated: '2026-04-12T18:24:56Z'
+generated_by: claude-opus-4-5-20251101
+workflow_run: 24313009897
+issue: 5251
+python_version: 3.14.3
+library_version: 6.0.0
+preview_url: https://storage.googleapis.com/pyplots-images/plots/scatter-lag/altair/plot.png
+preview_html: https://storage.googleapis.com/pyplots-images/plots/scatter-lag/altair/plot.html
+quality_score: 90
+review:
+  strengths:
+  - Viridis colormap with opacity=0.45 and white edge strokes creates a professional,
+    layered look
+  - 'All required spec features present: diagonal reference line, time-index color
+    coding, correlation annotation, configurable lag'
+  - Altair layer composition (reference_line + points + annotation) is idiomatic and
+    clean
+  - Explicit font sizing throughout ensures legibility at 4800x2700px
+  - Subtle grid (opacity 0.15), no axis border, and removed domain line give a polished
+    clean background
+  weaknesses:
+  - DE-01 held back by lacking a truly distinctive visual signature (e.g., custom
+    highlight color, marginal distribution element)
+  - DE-02 could reach 6/6 with more generous outer padding and intentional whitespace
+    strategy
+  - 'VQ-06 partial: axis labels use mathematical notation without physical units (appropriate
+    for abstract data but not 2/2)'
+  - 'LM-02: could leverage Altair''s selection/brush interactivity or faceting to
+    show multiple lag values side-by-side'
+  image_description: A scatter lag plot showing 499 points from a synthetic AR(1)
+    process (phi=0.85). The x-axis is labeled "y(t)" and the y-axis "y(t + 1)", both
+    ranging from about -5 to 6. Points are colored using the viridis colormap (purple/blue
+    = early time index, yellow-green = late time index), with opacity=0.45 and white
+    edge strokes. A dashed gray diagonal reference line (y=x) runs from bottom-left
+    to top-right, making the strong positive autocorrelation immediately visible.
+    A bold "r = 0.834" annotation appears in the lower-right of the plot area. The
+    title reads "scatter-lag · altair · pyplots.ai" with subtitle "AR(1) process (phi
+    = 0.85) | lag = 1". A vertical viridis color legend labeled "Time Index" is positioned
+    on the right. The background is white with an extremely subtle grid, no axis border,
+    and clean typography throughout.
+  criteria_checklist:
+    visual_quality:
+      score: 29
+      max: 30
+      items:
+      - id: VQ-01
+        name: Text Legibility
+        score: 8
+        max: 8
+        passed: true
+        comment: 'All font sizes explicitly set: title 28pt, subtitle 18pt, axis titles
+          22pt, tick labels 18pt, legend 16pt, annotation 20pt bold'
+      - id: VQ-02
+        name: No Overlap
+        score: 6
+        max: 6
+        passed: true
+        comment: No text or data element collisions. Legend well-separated from plot
+          area.
+      - id: VQ-03
+        name: Element Visibility
+        score: 6
+        max: 6
+        passed: true
+        comment: 500 points at size=45, opacity=0.45. Matches 300+ guideline. White
+          strokes help distinguish overlapping points.
+      - id: VQ-04
+        name: Color Accessibility
+        score: 4
+        max: 4
+        passed: true
+        comment: Viridis is perceptually uniform and colorblind-safe.
+      - id: VQ-05
+        name: Layout & Canvas
+        score: 4
+        max: 4
+        passed: true
+        comment: Plot fills ~70% of canvas with balanced margins. Well-distributed
+          elements.
+      - id: VQ-06
+        name: Axis Labels & Title
+        score: 1
+        max: 2
+        passed: true
+        comment: Labels y(t) and y(t+1) are descriptive but carry no physical units.
+    design_excellence:
+      score: 13
+      max: 20
+      items:
+      - id: DE-01
+        name: Aesthetic Sophistication
+        score: 5
+        max: 8
+        passed: true
+        comment: 'Above defaults: viridis colormap, opacity with white edge strokes,
+          dashed reference line, no border/domain lines. Cohesive but not publication-ready.'
+      - id: DE-02
+        name: Visual Refinement
+        score: 4
+        max: 6
+        passed: true
+        comment: Subtle grid (0.15 opacity), removed axis border, no domain line,
+          soft tick colors. More refined than defaults.
+      - id: DE-03
+        name: Data Storytelling
+        score: 4
+        max: 6
+        passed: true
+        comment: r=0.834 annotation quantifies the pattern; time coloring reveals
+          temporal structure; diagonal reference guides interpretation.
+    spec_compliance:
+      score: 15
+      max: 15
+      items:
+      - id: SC-01
+        name: Plot Type
+        score: 5
+        max: 5
+        passed: true
+        comment: 'Correct lag plot: scatter of y(t) vs y(t+k).'
+      - id: SC-02
+        name: Required Features
+        score: 4
+        max: 4
+        passed: true
+        comment: Diagonal reference line, color by time index, correlation annotation,
+          configurable lag variable all present.
+      - id: SC-03
+        name: Data Mapping
+        score: 3
+        max: 3
+        passed: true
+        comment: X=y(t), Y=y(t+1). Correct orientation. All 499 points visible.
+      - id: SC-04
+        name: Title & Legend
+        score: 3
+        max: 3
+        passed: true
+        comment: Title exactly 'scatter-lag · altair · pyplots.ai'. Legend labeled
+          'Time Index'.
+    data_quality:
+      score: 15
+      max: 15
+      items:
+      - id: DQ-01
+        name: Feature Coverage
+        score: 6
+        max: 6
+        passed: true
+        comment: 'Shows all key aspects: autocorrelation pattern, diagonal reference,
+          temporal coloring, correlation quantification.'
+      - id: DQ-02
+        name: Realistic Context
+        score: 5
+        max: 5
+        passed: true
+        comment: AR(1) process (phi=0.85) is a real, neutral statistical scenario.
+          Parameters clearly labeled in subtitle.
+      - id: DQ-03
+        name: Appropriate Scale
+        score: 4
+        max: 4
+        passed: true
+        comment: 500 observations with realistic AR(1) value range (~-5 to 6). Correlation
+          r=0.834 expected for phi=0.85.
+    code_quality:
+      score: 10
+      max: 10
+      items:
+      - id: CQ-01
+        name: KISS Structure
+        score: 3
+        max: 3
+        passed: true
+        comment: 'Linear: imports -> data -> chart layers -> save. No functions or
+          classes.'
+      - id: CQ-02
+        name: Reproducibility
+        score: 2
+        max: 2
+        passed: true
+        comment: np.random.seed(42) set.
+      - id: CQ-03
+        name: Clean Imports
+        score: 2
+        max: 2
+        passed: true
+        comment: altair, numpy, pandas — all used, none unnecessary.
+      - id: CQ-04
+        name: Code Elegance
+        score: 2
+        max: 2
+        passed: true
+        comment: Layer composition (reference_line + points + annotation) is clean
+          and idiomatic Altair.
+      - id: CQ-05
+        name: Output & API
+        score: 1
+        max: 1
+        passed: true
+        comment: Saves plot.png with scale_factor=3.0 and plot.html. Current altair
+          6.0.0 API.
+    library_features:
+      score: 8
+      max: 10
+      items:
+      - id: LM-01
+        name: Idiomatic Usage
+        score: 5
+        max: 5
+        passed: true
+        comment: 'Expert Altair: declarative encoding with typed fields (Q/N), layer
+          composition via +, alt.Scale/Axis/Legend, configure_* global styling.'
+      - id: LM-02
+        name: Distinctive Features
+        score: 3
+        max: 5
+        passed: true
+        comment: Layer composition, interactive tooltips, HTML export, declarative
+          viridis scheme. Distinctly Altair but basic within library capabilities.
+  verdict: APPROVED
+impl_tags:
+  dependencies: []
+  techniques:
+  - layer-composition
+  - annotations
+  - hover-tooltips
+  - html-export
+  patterns:
+  - data-generation
+  dataprep: []
+  styling:
+  - custom-colormap
+  - alpha-blending
+  - edge-highlighting