Add benchmarks for indexing and searching #109

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open

JJ-Pineda wants to merge 2 commits into main from add_benchmarks

benches/compound_processing_benches.rs

-Original file line number
+Diff line change
@@ -0,0 +1,49 @@
+    #![feature(test)]
+    use cheminee::search::compound_processing::*;
+    extern crate test;
+    use rdkit::ROMol;
+    use test::Bencher;
+    #[bench]
+    fn bench_process_cpd(b: &mut Bencher) {
+        let smiles1 =
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O";
+        b.iter(|| {
+            let _ = process_cpd(smiles1, false);
+        });
+    }
+    // running 1 test
+    // test bench_process_cpd ... bench:   5,527,616.65 ns/iter (+/- 337,547.33)
+    #[bench]
+    fn bench_standardize_mol(b: &mut Bencher) {
+        let mol = ROMol::from_smiles(
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O",
+        )
+        .unwrap();
+        b.iter(|| {
+            let _ = standardize_mol(&mol);
+        });
+    }
+    // running 1 test
+    // test bench_standardize_mol ... bench:   3,906,768.75 ns/iter (+/- 230,949.77)
+    #[bench]
+    fn bench_get_cpd_properties(b: &mut Bencher) {
+        let smiles1 =
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O";
+        let canon_taut = standardize_smiles(smiles1, false).unwrap();
+        b.iter(|| {
+            let _ = get_cpd_properties(&canon_taut);
+        });
+    }
+    // running 1 test
+    // test bench_get_cpd_properties ... bench:   1,329,194.77 ns/iter (+/- 183,436.78)

benches/fingerprint_benches.rs

-Original file line number
+Diff line change
@@ Expand Up / @@ -18,16 +18,31 @@ fn bench_tanimoto_distance(b: &mut Bencher) { @@
     }
     // running 1 test
-    // Before: test bench_tanimoto_similarity ... bench:     132,690 ns/iter (+/- 2,681)
-    // Now: test bench_tanimoto_distance ... bench:       2,653 ns/iter (+/- 790)
+    // test bench_tanimoto_distance ... bench:       2,520.20 ns/iter (+/- 151.54)
     #[bench]
     fn bench_fingerprint_generation(b: &mut Bencher) {
-        let smiles = "[N]Cc1cncc2c(=O)c3cccc(CCC(=O)O)c3[nH]c12";
+        let smiles =
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O";
         let romol = ROMol::from_smiles(smiles).unwrap();
         b.iter(|| romol.fingerprint());
     }
     // running 1 test
-    // test bench_fingerprint_generation ... bench:   1,161,160 ns/iter (+/- 274,184)
+    // test bench_fingerprint_generation ... bench:     784,123.44 ns/iter (+/- 137,752.50)
+    #[bench]
+    fn bench_fingerprint_exact_match(b: &mut Bencher) {
+        let mol1 = ROMol::from_smiles(
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O",
+        )
+        .unwrap();
+        let fp1 = mol1.fingerprint();
+        let fp2 = fp1.clone();
+        b.iter(|| fp1.0 == fp2.0);
+    }
+    // running 1 test
+    // test bench_fingerprint_exact_match ... bench:         372.45 ns/iter (+/- 30.19)

benches/indexing_benches.rs

-Original file line number
+Diff line change
@@ -0,0 +1,43 @@
+    #![feature(test)]
+    extern crate test;
+    use cheminee::command_line::indexing::bulk_index::create_tantivy_doc;
+    use cheminee::indexing::KNOWN_DESCRIPTORS;
+    use cheminee::schema::LIBRARY;
+    use serde_json::Value;
+    use std::collections::HashMap;
+    use tantivy::schema::Field;
+    use test::Bencher;
+    #[bench]
+    fn bench_create_tantivy_doc(b: &mut Bencher) {
+        let schema = LIBRARY.get("descriptor_v1").unwrap().clone();
+        let smiles_field = schema.get_field("smiles").unwrap();
+        let fingerprint_field = schema.get_field("fingerprint").unwrap();
+        let extra_data_field = schema.get_field("extra_data").unwrap();
+        let descriptor_fields = KNOWN_DESCRIPTORS
+            .iter()
+            .map(|kd| (*kd, schema.get_field(kd).unwrap()))
+            .collect::<HashMap<&str, Field>>();
+        let smi =
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O";
+        let mut map = serde_json::Map::new();
+        map.insert("smiles".to_string(), Value::String(smi.to_string()));
+        let record: Value = Value::Object(map);
+        b.iter(|| {
+            let _ = create_tantivy_doc(
+                record.clone(),
+                smiles_field,
+                fingerprint_field,
+                &descriptor_fields,
+                extra_data_field,
+            );
+        });
+    }
+    // running 1 test
+    // test bench_create_tantivy_doc ... bench:   6,527,306.20 ns/iter (+/- 562,411.21)

benches/scaffold_benches.rs

-Original file line number
+Diff line change
@@ Expand Up / @@ -13,3 +13,6 @@ fn bench_scaffold_search(b: &mut Bencher) { @@
         b.iter(|| scaffold_search(&fp.0, &mol, &PARSED_SCAFFOLDS));
     }
+    // running 1 test
+    // test bench_scaffold_search ... bench:     756,170.85 ns/iter (+/- 27,301.91)

benches/structure_matching_benches.rs

-Original file line number
+Diff line change
@@ -0,0 +1,80 @@
+    #![feature(test)]
+    extern crate test;
+    use cheminee::search::structure_matching::exact_match;
+    use rdkit::{substruct_match, ROMol, SubstructMatchParameters};
+    use test::Bencher;
+    #[bench]
+    fn bench_exact_match_no_chirality(b: &mut Bencher) {
+        let romol1 = ROMol::from_smiles(
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O",
+        )
+        .unwrap();
+        let romol2 = romol1.clone();
+        b.iter(|| {
+            let _ = exact_match(&romol1, &romol2, false);
+        });
+    }
+    // running 1 test
+    // test bench_exact_match_no_chirality ... bench:      10,117.06 ns/iter (+/- 423.84)
+    #[bench]
+    fn bench_exact_match_yes_chirality(b: &mut Bencher) {
+        let romol1 = ROMol::from_smiles(
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O",
+        )
+        .unwrap();
+        let romol2 = romol1.clone();
+        b.iter(|| {
+            let _ = exact_match(&romol1, &romol2, true);
+        });
+    }
+    // running 1 test
+    // test bench_exact_match_yes_chirality ... bench:      18,131.02 ns/iter (+/- 790.45)
+    #[bench]
+    fn bench_substructure_match_no_chirality(b: &mut Bencher) {
+        let super_mol = ROMol::from_smiles(
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O",
+        )
+        .unwrap();
+        let sub_mol =
+            ROMol::from_smiles("C[C@@H]1[C@H]([C@H]([C@@H](O1)[N]2C=NC3=C(N=CN=C23)N)O)O").unwrap();
+        let mut params = SubstructMatchParameters::default();
+        params.set_use_chirality(false);
+        b.iter(|| {
+            let _ = substruct_match(&super_mol, &sub_mol, &params);
+        });
+    }
+    // running 1 test
+    // test bench_substructure_match_no_chirality ... bench:       4,209.80 ns/iter (+/- 318.14)
+    #[bench]
+    fn bench_substructure_match_yes_chirality(b: &mut Bencher) {
+        let super_mol = ROMol::from_smiles(
+            "C[S+](CC[C@@H](C(=O)[O-])[NH3+])C[C@@H]1[C@H]([C@H]([C@@H](O1)N2C=NC3=C(N=CN=C32)N)O)O",
+        )
+        .unwrap();
+        let sub_mol =
+            ROMol::from_smiles("C[C@@H]1[C@H]([C@H]([C@@H](O1)[N]2C=NC3=C(N=CN=C23)N)O)O").unwrap();
+        let mut params = SubstructMatchParameters::default();
+        params.set_use_chirality(true);
+        b.iter(|| {
+            let _ = substruct_match(&super_mol, &sub_mol, &params);
+        });
+    }
+    // running 1 test
+    // test bench_substructure_match_yes_chirality ... bench:       7,313.00 ns/iter (+/- 532.24)

src/command_line/indexing/bulk_index.rs

-Original file line number
+Diff line change
@@ Expand Up / @@ -100,7 +100,7 @@ pub fn action(matches: &ArgMatches) -> eyre::Result<()> { @@
         Ok(())
     }
-    fn create_tantivy_doc(
+    pub fn create_tantivy_doc(
         record: serde_json::Value,
         smiles_field: Field,
         fingerprint_field: Field,
@@ Expand Down @@

src/search/identity_search.rs

-Original file line number
+Diff line change
@@ Expand Up / @@ -39,7 +39,7 @@ pub fn identity_search( @@
                     smiles_field,
                     fingerprint_field,
                     extra_data_field,
-                    &searcher,
+                    searcher,
                     &query_mol_mutex.lock().unwrap(),
                     query_fingerprint,
                     use_chirality,
@@ Expand Down @@

src/search/structure_search.rs

-Original file line number
+Diff line change
@@ Expand Up / @@ -68,7 +68,7 @@ pub fn structure_search( @@
                         smiles_field,
                         fingerprint_field,
                         extra_data_field,
-                        &searcher,
+                        searcher,
                         &query_mol_mutex.lock().unwrap(),
                         query_fingerprint,
                         method,
@@ Expand Down @@

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add benchmarks for indexing and searching #109

Uh oh!

Diff view

Diff view

There are no files selected for viewing

Uh oh!

Add benchmarks for indexing and searching #109

Are you sure you want to change the base?

Uh oh!

Add benchmarks for indexing and searching #109

Uh oh!

Uh oh!

Diff view

Diff view

There are no files selected for viewing

Uh oh!