Iteration 299: Add benchmarks for truncate, filter_labels, assign, transform_agg

Run: https://github.com/githubnext/tsessebe/actions/runs/25152235433

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

Author: github-actions[bot]

benchmarks/pandas/bench_assign.py

@@ -0,0 +1,28 @@
+"""Benchmark: DataFrame.assign — add computed columns to a 100k-row DataFrame"""
+import json, time
+import numpy as np
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+df = pd.DataFrame({
+    "a": np.arange(ROWS, dtype=float),
+    "b": np.arange(ROWS, dtype=float) * 2,
+})
+
+for _ in range(WARMUP):
+    df.assign(c=lambda d: d["a"] + d["b"])
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    df.assign(c=lambda d: d["a"] + d["b"])
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "assign",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_filter_labels.py

@@ -0,0 +1,31 @@
+"""Benchmark: DataFrame.filter by items on 100k-row DataFrame"""
+import json, time
+import numpy as np
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+df = pd.DataFrame({
+    "alpha": np.arange(ROWS, dtype=float),
+    "beta": np.arange(ROWS, dtype=float) * 2,
+    "gamma": np.arange(ROWS, dtype=float) * 3,
+    "delta": np.arange(ROWS, dtype=float) * 4,
+    "epsilon": np.arange(ROWS, dtype=float) * 5,
+})
+
+for _ in range(WARMUP):
+    df.filter(items=["alpha", "gamma", "epsilon"])
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    df.filter(items=["alpha", "gamma", "epsilon"])
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "filter_labels",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_transform_agg.py

@@ -0,0 +1,27 @@
+"""Benchmark: Series.transform — transform a 100k-element Series"""
+import json, time
+import numpy as np
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+data = (np.arange(ROWS) % 500) + 1.0
+idx = np.arange(ROWS) % 500
+s = pd.Series(data, index=idx)
+
+for _ in range(WARMUP):
+    s.transform("mean")
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.transform("mean")
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "transform_agg",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_truncate.py

@@ -0,0 +1,26 @@
+"""Benchmark: truncate on 100k-element Series"""
+import json, time
+import numpy as np
+import pandas as pd
+
+ROWS = 100_000
+WARMUP = 3
+ITERATIONS = 10
+
+data = np.arange(ROWS) * 0.5
+s = pd.Series(data, index=np.arange(ROWS))
+
+for _ in range(WARMUP):
+    s.truncate(before=10_000, after=90_000)
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.truncate(before=10_000, after=90_000)
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "truncate",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/tsb/bench_assign.ts

@@ -0,0 +1,36 @@
+/**
+ * Benchmark: dataFrameAssign — add computed columns to a 100k-row DataFrame
+ */
+import { DataFrame, dataFrameAssign } from "../../src/index.js";
+
+const ROWS = 100_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+
+const df = DataFrame.fromColumns({
+  a: Float64Array.from({ length: ROWS }, (_, i) => i),
+  b: Float64Array.from({ length: ROWS }, (_, i) => i * 2),
+});
+
+for (let i = 0; i < WARMUP; i++) {
+  dataFrameAssign(df, {
+    c: (d: DataFrame) => d.col("a").add(d.col("b")),
+  });
+}
+
+const start = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  dataFrameAssign(df, {
+    c: (d: DataFrame) => d.col("a").add(d.col("b")),
+  });
+}
+const total = performance.now() - start;
+
+console.log(
+  JSON.stringify({
+    function: "assign",
+    mean_ms: total / ITERATIONS,
+    iterations: ITERATIONS,
+    total_ms: total,
+  }),
+);

benchmarks/tsb/bench_filter_labels.ts

@@ -0,0 +1,36 @@
+/**
+ * Benchmark: filterDataFrame by items and regex on 100k-row DataFrame
+ */
+import { DataFrame, filterDataFrame } from "../../src/index.js";
+
+const ROWS = 100_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+
+const df = DataFrame.fromColumns({
+  alpha: Float64Array.from({ length: ROWS }, (_, i) => i),
+  beta: Float64Array.from({ length: ROWS }, (_, i) => i * 2),
+  gamma: Float64Array.from({ length: ROWS }, (_, i) => i * 3),
+  delta: Float64Array.from({ length: ROWS }, (_, i) => i * 4),
+  epsilon: Float64Array.from({ length: ROWS }, (_, i) => i * 5),
+});
+
+// Filter by items
+for (let i = 0; i < WARMUP; i++) {
+  filterDataFrame(df, { items: ["alpha", "gamma", "epsilon"] });
+}
+
+const start = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  filterDataFrame(df, { items: ["alpha", "gamma", "epsilon"] });
+}
+const total = performance.now() - start;
+
+console.log(
+  JSON.stringify({
+    function: "filter_labels",
+    mean_ms: total / ITERATIONS,
+    iterations: ITERATIONS,
+    total_ms: total,
+  }),
+);

benchmarks/tsb/bench_transform_agg.ts

@@ -0,0 +1,30 @@
+/**
+ * Benchmark: seriesTransform — transform a 100k-element Series
+ */
+import { Series, seriesTransform } from "../../src/index.js";
+
+const ROWS = 100_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+
+const data = Float64Array.from({ length: ROWS }, (_, i) => (i % 500) + 1);
+const s = new Series({ data, index: Array.from({ length: ROWS }, (_, i) => i % 500) });
+
+for (let i = 0; i < WARMUP; i++) {
+  seriesTransform(s, "mean");
+}
+
+const start = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  seriesTransform(s, "mean");
+}
+const total = performance.now() - start;
+
+console.log(
+  JSON.stringify({
+    function: "transform_agg",
+    mean_ms: total / ITERATIONS,
+    iterations: ITERATIONS,
+    total_ms: total,
+  }),
+);

benchmarks/tsb/bench_truncate.ts

@@ -0,0 +1,30 @@
+/**
+ * Benchmark: truncateSeries on 100k-element Series
+ */
+import { Series, truncateSeries } from "../../src/index.js";
+
+const ROWS = 100_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+
+const data = Float64Array.from({ length: ROWS }, (_, i) => i * 0.5);
+const s = new Series({ data, index: Array.from({ length: ROWS }, (_, i) => i) });
+
+for (let i = 0; i < WARMUP; i++) {
+  truncateSeries(s, 10_000, 90_000);
+}
+
+const start = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  truncateSeries(s, 10_000, 90_000);
+}
+const total = performance.now() - start;
+
+console.log(
+  JSON.stringify({
+    function: "truncate",
+    mean_ms: total / ITERATIONS,
+    iterations: ITERATIONS,
+    total_ms: total,
+  }),
+);