The existing notebooks covered basic functionality but lacked demonstrations of key ergonomic features added in v1.0: built-in datasets, dplyr-style operations, pipelines, uncertainty quantification, and the plugin system.

Changes

tutorial_10min.ipynb - Added 4 sections:

• Built-in datasets: load_aarhus_cs(), make_random_multilayer(), list_datasets()
• Dplyr-style graph ops: nodes().filter().mutate().group_by().summarise()
• Centrality uncertainty: Multiple centrality metrics (degree, betweenness, closeness, eigenvector) with bootstrap CI, stability analysis with Coefficient of Variation
• Pipeline API: Pipeline([ComputeStats(), FilterNodes(), LouvainCommunity()])

query_with_dsl.ipynb - Added 4 sections:

• Layer grouping: per_layer(), top_k(), group_summary()
• Enhanced uncertainty quantification: .uq(method="bootstrap", n_samples=50, ci=0.95) for multiple centrality metrics
• Uncertainty visualization: 4-panel plot showing error bars, CI widths, and CV comparison across centralities
• Temporal queries: .window(start=1.0, end=3.0)

simulate_dynamics.ipynb - Added 3 sections:

• DynamicsBuilder: D.process(SIR(...)).initial(...).steps(...).run()
• SimulationResult analysis: measure extraction, peak detection, attack rate
• Stochastic uncertainty: multiple runs with mean/std confidence bands

community_detection.ipynb - Added 3 sections:

• Null model testing: configuration_model() for significance
• Bootstrap uncertainty: CI for modularity and community count
• Plugin system: CommunityPlugin base class with PluginRegistry

Example: Centrality with Uncertainty

# Compute multiple centralities with uncertainty quantification
result = (
    Q.nodes()
     .from_layers(L["*"])
     .uq(method="bootstrap", n_samples=50, ci=0.95)
     .compute(
         "degree",
         "betweenness_centrality",
         "closeness_centrality",
         "eigenvector_centrality"
     )
     .order_by("-betweenness_centrality")
     .execute(network)
)

# Get results with uncertainty columns (_std, _ci95_low, _ci95_high)
df = result.to_pandas(expand_uncertainty=True)

# Analyze stability with Coefficient of Variation
df['betweenness_cv'] = df['betweenness_centrality_std'] / df['betweenness_centrality']
# CV < 0.2 indicates stable centrality ranking

Example: Dplyr-style Operations

# Before: Manual filtering and computation
nodes = network.get_nodes()
high_degree = [n for n in nodes if network.degree(n) > 2]

# After: Dplyr-style chainable operations with grouping
df = (
    nodes(network)
    .filter(lambda n: n["degree"] > 2)
    .mutate(importance=lambda n: n["degree"] / n["betweenness_centrality"])
    .group_by("layer")
    .summarise(avg_importance=("importance", np.mean))
    .to_pandas()
)

All notebooks retain valid JSON structure. Enhanced with detailed inline comments explaining bootstrap resampling, null models, fluent APIs, and centrality stability assessment.

• Fixes notebook demos #935

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.