Mostly fixed?

Experiments/Scripts/coloring_methods.jl

@@ -2,7 +2,7 @@
 using Profile
 include("../Experiments.jl")
 
-datasets = [human]
+datasets = [aids]
 partitioning_schemes = [
                         [(Degree, 64)],
                         [(NeighborNodeLabels, 64)],
@@ -21,7 +21,7 @@ for dataset in datasets
     end
 end
 
-build_experiments(experiment_params)
+#build_experiments(experiment_params)
 
 run_estimation_experiments(experiment_params)
 

Experiments/run_estimators.jl

@@ -6,7 +6,7 @@ function run_estimation_experiments(experiment_params_list::Vector{ExperimentPar
         !isfile(summary_file_location) && error("The summary has not been built yet! \n Attempted File Location: $(summary_file_location)")
         summary::ColorSummary = deserialize(summary_file_location)
         experiment_results = []
-        push!(experiment_results, ("UpperBound", "Estimate", "LowerBound", "TrueCard", "EstimationTime", "QueryType", "QueryPath"))
+        push!(experiment_results, ("Estimate", "TrueCard", "EstimationTime", "QueryType", "QueryPath"))
         for i in 1:length(all_queries[dataset])
             query::QueryGraph = all_queries[dataset][i].query
             query_path = all_queries[dataset][i].query_path
@@ -17,12 +17,9 @@ function run_estimation_experiments(experiment_params_list::Vector{ExperimentPar
                                     sampling_strategy=experiment_params.sampling_strategy,
                                     only_shortest_path_cycle= experiment_params.only_shortest_path_cycle)) for _ in 1:3]
             estimate_time = median([x.time for x in  estimate_results]) # Convert back to seconds from nano seconds
-            bounds = estimate_results[1].value
-            upper_bound = bounds[3]
-            estimate = max(1, bounds[2])
-            lower_bound = bounds[1]
+            estimate = estimate_results[1].value
             query_type = all_queries[dataset][i].query_type
-            push!(experiment_results, (upper_bound, estimate, lower_bound, exact_size, estimate_time, query_type, query_path))
+            push!(experiment_results, (estimate, exact_size, estimate_time, query_type, query_path))
         end
         results_file_location = "Experiments/Results/Estimation_"  * params_to_results_filename(experiment_params)
         writedlm(results_file_location, experiment_results, ",")

Source/ColorSummary.jl

@@ -335,7 +335,7 @@ function generate_color_summary(g::DataGraph, params::ColorSummaryParams=ColorSu
             for c1 in keys(color_to_color_edge_list[edge_label][vertex_label])
                 edge_deg[edge_label][vertex_label][c1] = Dict()
                 for c2 in keys(color_to_color_edge_list[edge_label][vertex_label][c1])
-                    edge_deg[edge_label][vertex_label][c1][c2] = DS(g, color_to_color_edge_list[edge_label][vertex_label][c1][c2])
+                    edge_deg[edge_label][vertex_label][c1][c2] = DS(g, color_to_color_edge_list[edge_label][vertex_label][c1][c2], color_label_cardinality[c1][-1])
                 end
             end
         end

Source/DegreeStats.jl

@@ -76,7 +76,7 @@ mutable struct MinDegStats <:DegreeStats
     min_out::Float64
 end
 
-function MinDegStats(g::DataGraph, edges::Vector{Tuple{NodeId, NodeId, Bool}})
+function MinDegStats(g::DataGraph, edges::Vector{Tuple{NodeId, NodeId, Bool}}, color_size::Int)
     length(edges) == 0 && return MinDegStats(0,0)
     in_counter = counter(NodeId)
     out_counter = counter(NodeId)
@@ -104,7 +104,7 @@ end
 
 struct MinAccumulator <:StatAccumulator
     weight::Float64
-    MinAccumulator(c::Float64) = new(c)
+    MinAccumulator(c::Union{Float64, Int64}) = new(c)
 end
 get_count(w::MinAccumulator) = w.weight
 sum_colorings(w1::MinAccumulator, w2::MinAccumulator) = MinAccumulator(w1.weight + w2.weight)
@@ -137,23 +137,19 @@ mutable struct AvgDegStats <:DegreeStats
     avg_out::Float64
 end
 
-function AvgDegStats(g::DataGraph, edges::Vector{Tuple{NodeId, NodeId, Bool}})
+function AvgDegStats(g::DataGraph, edges::Vector{Tuple{NodeId, NodeId, Bool}}, color_size::Int)
     length(edges) == 0 && return AvgDegStats(0,0)
     in_counter = counter(NodeId)
     out_counter = counter(NodeId)
-    unique_nodes = Set()
     for edge in edges
         if edge[3]
             inc!(out_counter, edge[1])
-            push!(unique_nodes, edge[1])
         else
             inc!(in_counter, edge[1])
-            push!(unique_nodes, edge[1])
         end
-
     end
-    avg_in = sum([x for x in values(in_counter)]; init=0)/length(unique_nodes)
-    avg_out = sum([x for x in values(out_counter)]; init=0)/length(unique_nodes)
+    avg_in = sum([x for x in values(in_counter)]; init=0)/color_size
+    avg_out = sum([x for x in values(out_counter)]; init=0)/color_size
     return AvgDegStats(avg_in, avg_out)
 end
 get_in_deg_estimate(d::AvgDegStats) = d.avg_in
@@ -169,7 +165,7 @@ end
 
 struct AvgAccumulator <:StatAccumulator
     weight::Float64
-    AvgAccumulator(c::Float64) = new(c)
+    AvgAccumulator(c::Union{Float64, Int64}) = new(c)
 end
 
 get_count(w::AvgAccumulator) = w.weight
@@ -191,7 +187,7 @@ mutable struct MaxDegStats <:DegreeStats
     max_out::Float64
 end
 
-function MaxDegStats(g::DataGraph, edges::Vector{Tuple{NodeId, NodeId, Bool}})
+function MaxDegStats(g::DataGraph, edges::Vector{Tuple{NodeId, NodeId, Bool}}, color_size::Int)
     length(edges) == 0 && return MaxDegStats(0,0)
     in_counter = counter(NodeId)
     out_counter = counter(NodeId)
@@ -218,7 +214,7 @@ end
 
 struct MaxAccumulator <:StatAccumulator
     weight::Float64
-    MaxAccumulator(c::Float64) = new(c)
+    MaxAccumulator(c::Union{Float64, Int64}) = new(c)
 end
 get_count(w::MaxAccumulator) = w.weight
 sum_colorings(w1::MaxAccumulator, w2::MaxAccumulator) = MaxAccumulator(w1.weight + w2.weight)

Source/QuasiStableCardinalityEstimator.jl

@@ -24,7 +24,7 @@
     end
     deleteat!(current_query_nodes, nodeIdx)
     partial_paths = zeros(Color, length(current_query_nodes), length(keys(new_partial_paths)))
-    partial_weights = fill(W(), length(keys(new_partial_paths)))
+    partial_weights = fill(W(0.0), length(keys(new_partial_paths)))
 
     path_idx = 1
     for path in keys(new_partial_paths)
@@ -43,7 +43,7 @@ function sample_paths(partial_paths::Matrix{Color}, partial_weights::Vector{W},
     # if we want to sample more paths than there are existing nonzero paths,
     # then just return the original partial paths
     new_partial_paths = zeros(Color, size(partial_paths))
-    new_partial_weights = fill(W(), size(partial_weights))
+    new_partial_weights = fill(W(0.0), size(partial_weights))
     new_path_idx = 1
     for i in eachindex(partial_weights)
         if get_count(partial_weights[i]) > 0
@@ -61,7 +61,7 @@ function sample_paths(partial_paths::Matrix{Color}, partial_weights::Vector{W},
     end
 
     # sum up all of the bounds
-    overall_bounds_sum::Float64 = 0.0
+    overall_bounds_sum = 0.0
     for w in new_partial_weights
         overall_bounds_sum += get_count(w)
     end
@@ -75,7 +75,7 @@ function sample_paths(partial_paths::Matrix{Color}, partial_weights::Vector{W},
     sample_indices::Vector{Int} = sample(1:length(new_partial_weights), sample_weights,  num_samples; replace=false)
 
     # sum up the sampled bounds
-    sampled_bounds_sum::Float64 = 0
+    sampled_bounds_sum = 0
     for idx in sample_indices
         sampled_bounds_sum += get_count(new_partial_weights[idx])
     end
@@ -92,7 +92,7 @@ function sample_paths(partial_paths::Matrix{Color}, partial_weights::Vector{W},
         else
             1.0 / sample_weights[i]
         end
-        sampled_partial_weights[i] = scale_colorings(new_partial_weights[idx], inverse_sampling_probability)
+        sampled_partial_weights[i] = scale_coloring(new_partial_weights[idx], inverse_sampling_probability)
     end
     return sampled_partial_paths, sampled_partial_weights
 end
@@ -427,5 +427,8 @@ function get_cardinality_bounds(query::QueryGraph, summary::ColorSummary{DS}; ma
     for w in partial_weights
         final_estimate += get_count(w)
     end
+    if final_estimate == 0
+        println("ISSUE")
+    end
     return final_estimate
 end