From 37434f5baf632e839dc14b3c7d8745287cb9368a Mon Sep 17 00:00:00 2001
From: Douglas Rumbaugh <dbr4@psu.edu>
Date: Tue, 25 Jul 2023 11:17:36 -0400
Subject: Benchmarks: mtree and vptree benchmark updates

Note: cosine similarity doesn't seem to work for VPTree--I don't think
that it is actually a metric, upon further research. At the very least I
can't find anyone claiming it is, and I've found several people claiming
it isn't. On testing with the Word2Vec data, Euclidean distance works
insofar as the M-Tree and VPTree return the same KNN results for test
queries, whereas Cosine Similarity does not work.
---
 benchmarks/include/bench.h         |  3 ++-
 benchmarks/include/bench_utility.h | 15 +++++++++++++--
 2 files changed, 15 insertions(+), 3 deletions(-)

(limited to 'benchmarks/include')

diff --git a/benchmarks/include/bench.h b/benchmarks/include/bench.h
index 12d0a7e..586ff12 100644
--- a/benchmarks/include/bench.h
+++ b/benchmarks/include/bench.h
@@ -85,7 +85,7 @@ static bool insert_tput_bench(DE &de_index, std::fstream &file, size_t insert_cn
 }
 
 template <typename DE, de::RecordInterface R, typename QP, bool PROGRESS=true>
-static bool query_latency_bench(DE &de_index, std::vector<QP> queries, size_t trial_cnt=100) {
+static bool query_latency_bench(DE &de_index, std::vector<QP> queries, size_t trial_cnt=1) {
     char progbuf[25];
     if constexpr (PROGRESS) {
         sprintf(progbuf, "querying:");
@@ -102,6 +102,7 @@ static bool query_latency_bench(DE &de_index, std::vector<QP> queries, size_t tr
         auto start = std::chrono::high_resolution_clock::now();
         for (size_t j=0; j<queries.size(); j++) {
             auto res = de_index.query(&queries[j]);
+
             total_results += res.size();
         }
         auto stop = std::chrono::high_resolution_clock::now();
diff --git a/benchmarks/include/bench_utility.h b/benchmarks/include/bench_utility.h
index 6610ab4..28040be 100644
--- a/benchmarks/include/bench_utility.h
+++ b/benchmarks/include/bench_utility.h
@@ -40,7 +40,7 @@ typedef de::WeightedRecord<key_type, value_type, weight_type> WRec;
 typedef de::Record<key_type, value_type> Rec;
 
 const size_t W2V_SIZE = 300;
-typedef de::CosinePoint<double, W2V_SIZE> Word2VecRec;
+typedef de::EuclidPoint<double, W2V_SIZE> Word2VecRec;
 
 typedef de::DynamicExtension<WRec, de::WSS<WRec>, de::WSSQuery<WRec>> ExtendedWSS;
 typedef de::DynamicExtension<Rec, de::TrieSpline<Rec>, de::TrieSplineRangeQuery<Rec>> ExtendedTSRQ;
@@ -68,6 +68,17 @@ struct btree_key_extract {
     }
 };
 
+struct euclidean_distance {
+    double operator()(const Word2VecRec &first, const Word2VecRec &second) const {
+        double dist = 0;
+        for (size_t i=0; i<W2V_SIZE; i++) {
+            dist += (first.data[i] - second.data[i]) * (first.data[i] - second.data[i]);
+        }
+        
+        return std::sqrt(dist);
+    }
+};
+
 
 struct cosine_similarity {
     double operator()(const Word2VecRec &first, const Word2VecRec &second) const {
@@ -86,7 +97,7 @@ struct cosine_similarity {
 };
 
 typedef tlx::BTree<key_type, btree_record, btree_key_extract> TreeMap;
-typedef mt::mtree<Word2VecRec, cosine_similarity> MTree;
+typedef mt::mtree<Word2VecRec, euclidean_distance> MTree;
 
 static gsl_rng *g_rng;
 static std::set<WRec> *g_to_delete;
-- 
cgit v1.2.3