Implemented two methods (loop_vec and online)

uwdb · Nov 16, 2023 · 51c1d28 · 51c1d28
1 parent 7c5ea5a
commit 51c1d28
Show file tree

Hide file tree

Showing 5 changed files with 241 additions and 83 deletions.
diff --git a/Experiments/Scripts/sampling_strategies.jl b/Experiments/Scripts/sampling_strategies.jl
@@ -2,19 +2,24 @@
 using Profile
 include("../Experiments.jl")
 
-datasets = [human, aids, yeast, hprd, dblp, wordnet]
+#datasets = [human, aids, yeast, hprd, dblp, wordnet]
+datasets = [aids]
 
 experiment_params = Vector{ExperimentParams}()
 build_params = Vector{ExperimentParams}()
+num_colors = 128
 for dataset in datasets
-    push!(build_params, ExperimentParams(dataset=dataset))
-    for sampling_strategy in instances(SAMPLING_STRATEGY)
-        push!(experiment_params, ExperimentParams(dataset=dataset, sampling_strategy=sampling_strategy, inference_max_paths=128))
+    push!(build_params, ExperimentParams(dataset=dataset,  num_colors=num_colors))
+#    for sampling_strategy in instances(SAMPLING_STRATEGY)
+    for sampling_strategy in [redistributive, loop_vec, online]
+            push!(experiment_params, ExperimentParams(dataset=dataset, num_colors=num_colors, sampling_strategy=sampling_strategy, inference_max_paths=512))
     end
 end
 
-build_experiments(build_params)
+#build_experiments(build_params)
 
 run_estimation_experiments(experiment_params)
 
-graph_grouped_box_plot(experiment_params; grouping=sampling_type, filename="sampling_strategies_comparison")
+graph_grouped_box_plot(experiment_params; grouping=sampling_type, filename="sampling_strategies_comparison_error")
+
+graph_grouped_box_plot(experiment_params; y_type = runtime, grouping=sampling_type, filename="sampling_strategies_comparison_runtime")
diff --git a/Experiments/run_estimators.jl b/Experiments/run_estimators.jl
@@ -10,6 +10,7 @@ function run_estimation_experiments(experiment_params_list::Vector{ExperimentPar
         for i in 1:length(all_queries[dataset])
             query::QueryGraph = all_queries[dataset][i].query
             query_path = all_queries[dataset][i].query_path
+            println(query_path)
             exact_size = all_queries[dataset][i].exact_size
             estimate_results = [(@timed get_cardinality_bounds(query, summary;
                                     max_partial_paths = experiment_params.inference_max_paths,

diff --git a/Project.toml b/Project.toml
@@ -12,10 +12,12 @@ DataStructures = "864edb3b-99cc-5e75-8d2d-829cb0a9cfe8"
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
 GraphRecipes = "bd48cda9-67a9-57be-86fa-5b3c104eda73"
 Graphs = "86223c79-3864-5bf0-83f7-82e725a168b6"
+LoopVectorization = "bdcacae8-1622-11e9-2a5c-532679323890"
 Parquet2 = "98572fba-bba0-415d-956f-fa77e587d26d"
 Plots = "91a5bcdd-55d7-5caf-9e0b-520d859cae80"
 Probably = "2172800d-0309-5a57-a84f-d50c94757422"
 QuasiStableColors = "9c3856af-3e7c-4d34-a6af-a406867b22e4"
+QuickHeaps = "30b38841-0f52-47f8-a5f8-18d5d4064379"
 Revise = "295af30f-e4ad-537b-8983-00126c2a3abe"
 StatsBase = "2913bbd2-ae8a-5f71-8c99-4fb6c76f3a91"
 StatsPlots = "f3b207a7-027a-5e70-b257-86293d7955fd"

diff --git a/Source/CardinalityWithColors.jl b/Source/CardinalityWithColors.jl
@@ -1,8 +1,12 @@
 using Probably: BloomFilter, constrain
-using DataStructures: counter, Dict, Set, Vector, inc!, Queue
+using DataStructures: counter, inc!, Queue
+using QuickHeaps: PriorityQueue, dequeue!, enqueue!
 using AutoHashEquals
 using QuasiStableColors
 using StatsBase
+using Random
+using LoopVectorization
+using Distributions
 
 QSC = QuasiStableColors
 using Graphs: SimpleDiGraphFromIterator, Edge, DiGraph, edges, nv, ne, add_edge!,