fixed cycle prob update slightly

uwdb · Dec 7, 2023 · e27140d · e27140d
1 parent 4593325
commit e27140d
Show file tree

Hide file tree

Showing 5 changed files with 5 additions and 9 deletions.
diff --git a/Source/CardinalityWithColors.jl b/Source/CardinalityWithColors.jl
@@ -8,7 +8,6 @@ QSC = QuasiStableColors
 using Graphs: SimpleDiGraphFromIterator, Edge, DiGraph, edges, nv, ne, add_edge!,
                 add_vertex!, vertices, all_neighbors, src, dst, outneighbors, inneighbors
 
-
 BoolPath = Vector{Bool}
 NodeId = Int
 Color = Int16
@@ -72,8 +71,6 @@ function params_to_string(params::ColorSummaryParams)
     return summary_name
 end
 
-
-
 include("PropertyGraph.jl")
 include("datasets.jl")
 include("utils.jl")

diff --git a/Source/ColorSummary.jl b/Source/ColorSummary.jl
@@ -198,7 +198,6 @@ function update_edge_degrees!(summary::ColorSummary{AvgDegStats}, start_node, en
     summary.total_added_edges += 1
 end
 
-
 function generate_color_summary(g::DataGraph, params::ColorSummaryParams=ColorSummaryParams(); verbose=0, timing_vec::Vector{Float64} = Float64[], use_cycle_join_table=true)
     DS = params.deg_stats_type
     if (verbose > 0)

diff --git a/Source/DegreeStats.jl b/Source/DegreeStats.jl
@@ -1,7 +1,6 @@
 # This file defines the interface and several instantiations of degree statistics. These are
 # the core statistic that is used in the estimation procedure.
 
-
 abstract type DegreeStats end
 
 # A degree statistic is derived from a set of edges. We include the data graph as well in
@@ -16,6 +15,7 @@ end
 function get_out_deg_estimate(d::DegreeStats)
     throw(ErrorException("DegreeStats is an abstract type, you probably meant to call a particular instance."))
 end
+
 function get_in_deg_estimate(d::DegreeStats)
     throw(ErrorException("DegreeStats is an abstract type, you probably meant to call a particular instance."))
 end
@@ -136,7 +136,6 @@ end
 get_in_deg_estimate(d::AvgDegStats) = d.avg_in
 get_out_deg_estimate(d::AvgDegStats) = d.avg_out
 
-
 struct AvgAccumulator <:StatAccumulator
     weight::Float32
 end
@@ -147,9 +146,9 @@ scale_coloring(w::AvgAccumulator, s) = AvgAccumulator(w.weight * s)
 
 function extend_coloring(w::AvgAccumulator, d::AvgDegStats, out_edge::Bool)
     if out_edge
-        return AvgAccumulator(w.weight*d.avg_out)
+        return scale_coloring(w, d.avg_out)
     else
-        return AvgAccumulator(w.weight*d.avg_in)
+        return scale_coloring(w, d.avg_in)
     end
 end
 

diff --git a/Source/PropertyGraph.jl b/Source/PropertyGraph.jl
@@ -86,7 +86,7 @@ end
 
 function add_labeled_node!(g::DataGraph, node_labels=[], data_label=-1)
     add_vertex!(g.graph)
-    g.vertex_labels = [g.vertex_labels..., node_labels]
+    push!(g.vertex_labels, node_labels)
 end
 
 function print_graph!(g::PropertyGraph)

diff --git a/Source/QuasiStableCardinalityEstimator.jl b/Source/QuasiStableCardinalityEstimator.jl
@@ -251,6 +251,7 @@ function handle_extra_edges!(query::QueryGraph, summary::ColorSummary{DS}, parti
                     probability_no_edge *= 1.0 - get_independent_cycle_likelihood(edge_label, child_label, parent_color, child_color, summary)
                 end
             end
+            probability_no_edge *= 1.0 - summary.total_added_edges/summary.total_nodes^2
             partial_weights[i] = scale_coloring(partial_weights[i], (1.0 - probability_no_edge))
         end
     end
-Original file line number
+Diff line change
@@ Expand Up @@
                         probability_no_edge *= 1.0 - get_independent_cycle_likelihood(edge_label, child_label, parent_color, child_color, summary)
                     end
                 end
+                probability_no_edge *= 1.0 - summary.total_added_edges/summary.total_nodes^2
                 partial_weights[i] = scale_coloring(partial_weights[i], (1.0 - probability_no_edge))
             end
         end
@@ Expand Down @@