Iteration 115: Add 6 new benchmark pairs (340 total, +6 vs best 334)

Run: https://github.com/githubnext/tsessebe/actions/runs/24483302039

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Author: github-actions[bot]

benchmarks/pandas/bench_dt_is_quarter_start_end.py

@@ -0,0 +1,27 @@
+"""Benchmark: dt_is_quarter_start_end — is_quarter_start, is_quarter_end on 100k datetime values"""
+import json, time
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+dates = pd.date_range("2024-01-01", periods=ROWS, freq="1D")
+s = pd.Series(dates)
+
+for _ in range(WARMUP):
+    s.dt.is_quarter_start
+    s.dt.is_quarter_end
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.dt.is_quarter_start
+    s.dt.is_quarter_end
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "dt_is_quarter_start_end",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_dt_year_month_day.py

@@ -0,0 +1,29 @@
+"""Benchmark: dt_year_month_day — dt.year, dt.month, dt.day on 100k datetime values"""
+import json, time
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+dates = pd.date_range("2024-01-01", periods=ROWS, freq="1D")
+s = pd.Series(dates)
+
+for _ in range(WARMUP):
+    s.dt.year
+    s.dt.month
+    s.dt.day
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.dt.year
+    s.dt.month
+    s.dt.day
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "dt_year_month_day",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_ewm_var.py

@@ -0,0 +1,26 @@
+"""Benchmark: EWM var (span=20) on 100k-element Series"""
+import json, time
+import numpy as np
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+data = np.sin(np.arange(ROWS) * 0.01)
+s = pd.Series(data)
+
+for _ in range(WARMUP):
+    s.ewm(span=20).var()
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.ewm(span=20).var()
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "ewm_var",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_multi_index_duplicated.py

@@ -0,0 +1,31 @@
+"""Benchmark: MultiIndex.duplicated() and drop_duplicates() on 100k-pair MultiIndex"""
+import json, time
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+# Create a MultiIndex with duplicates (10k unique pairs repeated 10 times)
+a = [f"a{i % 100}" for i in range(ROWS)]
+b = [i % 1000 for i in range(ROWS)]
+tuples = list(zip(a, b))
+
+mi = pd.MultiIndex.from_tuples(tuples)
+
+for _ in range(WARMUP):
+    mi.duplicated()
+    mi.drop_duplicates()
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    mi.duplicated()
+    mi.drop_duplicates()
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "multi_index_duplicated",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_multi_index_isna_dropna.py

@@ -0,0 +1,34 @@
+"""Benchmark: MultiIndex.isna(), notna(), dropna() on 100k-pair MultiIndex with some nulls"""
+import json, time
+import pandas as pd
+import numpy as np
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+# Create a MultiIndex with some null values
+a = [None if i % 10 == 0 else f"a{i % 100}" for i in range(ROWS)]
+b = [None if i % 20 == 0 else i % 1000 for i in range(ROWS)]
+tuples = list(zip(a, b))
+
+mi = pd.MultiIndex.from_tuples(tuples)
+
+for _ in range(WARMUP):
+    mi.isna()
+    mi.notna()
+    mi.dropna()
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    mi.isna()
+    mi.notna()
+    mi.dropna()
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "multi_index_isna_dropna",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_multi_index_sort_equals.py

@@ -0,0 +1,31 @@
+"""Benchmark: MultiIndex sort_values and equals on 100k-pair MultiIndex"""
+import json, time
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+a = [f"a{i % 100}" for i in range(ROWS)]
+b = [i % 1000 for i in range(ROWS)]
+tuples = list(zip(a, b))
+
+mi = pd.MultiIndex.from_tuples(tuples)
+mi2 = pd.MultiIndex.from_tuples(tuples[:])
+
+for _ in range(WARMUP):
+    mi.sort_values()
+    mi.equals(mi2)
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    mi.sort_values()
+    mi.equals(mi2)
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "multi_index_sort_equals",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/tsb/bench_dt_is_quarter_start_end.ts

@@ -0,0 +1,33 @@
+/**
+ * Benchmark: dt_is_quarter_start_end — is_quarter_start, is_quarter_end on 100k datetime values
+ */
+import { Series } from "../../src/index.ts";
+
+const ROWS = 100_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+
+const now = new Date("2024-01-01").getTime();
+const data = Array.from({ length: ROWS }, (_, i) => new Date(now + i * 24 * 3600 * 1000));
+const s = new Series({ data });
+
+for (let i = 0; i < WARMUP; i++) {
+  s.dt.is_quarter_start();
+  s.dt.is_quarter_end();
+}
+
+const start = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  s.dt.is_quarter_start();
+  s.dt.is_quarter_end();
+}
+const total = performance.now() - start;
+
+console.log(
+  JSON.stringify({
+    function: "dt_is_quarter_start_end",
+    mean_ms: total / ITERATIONS,
+    iterations: ITERATIONS,
+    total_ms: total,
+  }),
+);

benchmarks/tsb/bench_dt_year_month_day.ts

@@ -0,0 +1,35 @@
+/**
+ * Benchmark: dt_year_month_day — dt.year(), dt.month(), dt.day() on 100k datetime values
+ */
+import { Series } from "../../src/index.ts";
+
+const ROWS = 100_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+
+const now = new Date("2024-01-01").getTime();
+const data = Array.from({ length: ROWS }, (_, i) => new Date(now + i * 24 * 3600 * 1000));
+const s = new Series({ data });
+
+for (let i = 0; i < WARMUP; i++) {
+  s.dt.year();
+  s.dt.month();
+  s.dt.day();
+}
+
+const start = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  s.dt.year();
+  s.dt.month();
+  s.dt.day();
+}
+const total = performance.now() - start;
+
+console.log(
+  JSON.stringify({
+    function: "dt_year_month_day",
+    mean_ms: total / ITERATIONS,
+    iterations: ITERATIONS,
+    total_ms: total,
+  }),
+);

benchmarks/tsb/bench_ewm_var.ts

@@ -0,0 +1,30 @@
+/**
+ * Benchmark: EWM var (span=20) on 100k-element Series
+ */
+import { Series } from "../../src/index.ts";
+
+const ROWS = 100_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+
+const data = Float64Array.from({ length: ROWS }, (_, i) => Math.sin(i * 0.01));
+const s = new Series(data);
+
+for (let i = 0; i < WARMUP; i++) {
+  s.ewm({ span: 20 }).var();
+}
+
+const start = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  s.ewm({ span: 20 }).var();
+}
+const total = performance.now() - start;
+
+console.log(
+  JSON.stringify({
+    function: "ewm_var",
+    mean_ms: total / ITERATIONS,
+    iterations: ITERATIONS,
+    total_ms: total,
+  }),
+);

benchmarks/tsb/bench_multi_index_duplicated.ts

@@ -0,0 +1,35 @@
+/**
+ * Benchmark: MultiIndex.duplicated() and dropDuplicates() on 100k-pair MultiIndex
+ */
+import { MultiIndex } from "../../src/index.ts";
+
+const ROWS = 100_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+
+// Create a MultiIndex with duplicates (10k unique pairs repeated 10 times)
+const a = Array.from({ length: ROWS }, (_, i) => `a${i % 100}`);
+const b = Array.from({ length: ROWS }, (_, i) => i % 1000);
+const tuples: [string, number][] = a.map((v, i) => [v, b[i]]);
+const mi = new MultiIndex({ tuples });
+
+for (let i = 0; i < WARMUP; i++) {
+  mi.duplicated();
+  mi.dropDuplicates();
+}
+
+const start = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  mi.duplicated();
+  mi.dropDuplicates();
+}
+const total = performance.now() - start;
+
+console.log(
+  JSON.stringify({
+    function: "multi_index_duplicated",
+    mean_ms: total / ITERATIONS,
+    iterations: ITERATIONS,
+    total_ms: total,
+  }),
+);