Merge pull request #311 from githubnext/autoloop/perf-comparison

[Autoloop: perf-comparison] Iteration 315: add to_markdown / to_latex…

Author: mrjf

benchmarks/pandas/bench_str_findall.py

@@ -0,0 +1,33 @@
+"""
+Benchmark: str.findall, str.extract (first match), str.count on 10k-element string Series
+"""
+import json
+import time
+import pandas as pd
+
+ROWS = 10_000
+WARMUP = 3
+ITERATIONS = 10
+
+data = [f"item{i} code{i * 3} ref{i + 1}" for i in range(ROWS)]
+s = pd.Series(data)
+pat = r"\d+"
+
+for _ in range(WARMUP):
+    s.str.findall(pat)
+    s.str.extract(r"(\d+)", expand=False)
+    s.str.count(pat)
+
+start = time.perf_counter()
+for _ in range(ITERATIONS):
+    s.str.findall(pat)
+    s.str.extract(r"(\d+)", expand=False)
+    s.str.count(pat)
+total = (time.perf_counter() - start) * 1000
+
+print(json.dumps({
+    "function": "str_findall",
+    "mean_ms": total / ITERATIONS,
+    "iterations": ITERATIONS,
+    "total_ms": total,
+}))

benchmarks/pandas/bench_to_markdown.py

@@ -0,0 +1,36 @@
+"""Benchmark: to_markdown and to_latex on a 1000-row DataFrame"""
+import json, time
+import numpy as np
+import pandas as pd
+
+ROWS = 1_000
+WARMUP = 3
+ITERATIONS = 10
+
+a = np.arange(ROWS) * 1.5
+b = [f"item_{i % 50}" for i in range(ROWS)]
+c = np.arange(ROWS) % 100
+df = pd.DataFrame({"a": a, "b": b, "c": c})
+
+for _ in range(WARMUP):
+    df.to_markdown()
+    df.to_latex()
+
+start_md = time.perf_counter()
+for _ in range(ITERATIONS):
+    df.to_markdown()
+total_md = (time.perf_counter() - start_md) * 1000
+
+start_ltx = time.perf_counter()
+for _ in range(ITERATIONS):
+    df.to_latex()
+total_ltx = (time.perf_counter() - start_ltx) * 1000
+
+total = total_md + total_ltx
+
+print(json.dumps({
+    "function": "to_markdown_latex",
+    "mean_ms": total / (ITERATIONS * 2),
+    "iterations": ITERATIONS * 2,
+    "total_ms": total,
+}))

benchmarks/tsb/bench_str_findall.ts

@@ -0,0 +1,34 @@
+/**
+ * Benchmark: strFindall, strFindFirst, strFindallCount on 10k-element string Series
+ */
+import { Series, strFindall, strFindFirst, strFindallCount } from "../../src/index.js";
+
+const ROWS = 10_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+const data = Array.from({ length: ROWS }, (_, i) => `item${i} code${i * 3} ref${i + 1}`);
+const s = new Series({ data });
+const pat = /\d+/g;
+
+for (let i = 0; i < WARMUP; i++) {
+  strFindall(s, pat);
+  strFindFirst(s, pat);
+  strFindallCount(s, pat);
+}
+
+const start = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  strFindall(s, pat);
+  strFindFirst(s, pat);
+  strFindallCount(s, pat);
+}
+const total = performance.now() - start;
+
+console.log(
+  JSON.stringify({
+    function: "str_findall",
+    mean_ms: total / ITERATIONS,
+    iterations: ITERATIONS,
+    total_ms: total,
+  }),
+);

benchmarks/tsb/bench_to_markdown.ts

@@ -0,0 +1,41 @@
+/**
+ * Benchmark: toMarkdown and toLaTeX on a 1000-row DataFrame
+ */
+import { DataFrame, toMarkdown, toLaTeX } from "../../src/index.js";
+
+const ROWS = 1_000;
+const WARMUP = 3;
+const ITERATIONS = 10;
+
+const a = Float64Array.from({ length: ROWS }, (_, i) => i * 1.5);
+const b = Array.from({ length: ROWS }, (_, i) => `item_${i % 50}`);
+const c = Int32Array.from({ length: ROWS }, (_, i) => i % 100);
+const df = DataFrame.fromColumns({ a, b, c });
+
+for (let i = 0; i < WARMUP; i++) {
+  toMarkdown(df);
+  toLaTeX(df);
+}
+
+const startMd = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  toMarkdown(df);
+}
+const totalMd = performance.now() - startMd;
+
+const startLtx = performance.now();
+for (let i = 0; i < ITERATIONS; i++) {
+  toLaTeX(df);
+}
+const totalLtx = performance.now() - startLtx;
+
+const total = totalMd + totalLtx;
+
+console.log(
+  JSON.stringify({
+    function: "to_markdown_latex",
+    mean_ms: total / (ITERATIONS * 2),
+    iterations: ITERATIONS * 2,
+    total_ms: total,
+  }),
+);