docs: update changelog

Author: gusye1234

Changelog.md

@@ -1,8 +1,8 @@
-### [0.0.37] - unreleased
+### [0.0.37] - 2025/7/15
 
 **breaking changes**:
 
-- Add event gist table, need migration
+- Add event gist table, need DB migration
 
 **Added** 
 

docs/experiments/locomo-benchmark/compute_p95_latency.py

@@ -0,0 +1,87 @@
+#!/usr/bin/env python3
+"""
+P95 Search Time Latency Demo Program
+
+This program simulates search operations and computes the 95th percentile (p95) latency.
+It includes timing measurements, data generation, and statistical analysis.
+"""
+
+import time
+import random
+import numpy as np
+import statistics
+from typing import List, Dict, Any
+from dataclasses import dataclass
+from concurrent.futures import ThreadPoolExecutor
+import json
+
+
+class LatencyAnalyzer:
+    """Analyzes search latencies and computes statistics."""
+
+    def __init__(self, results: list[float]):
+        self.results: List[float] = results
+
+    def get_latencies(self) -> List[float]:
+        """Extract latency values from all results."""
+        return self.results
+
+    def compute_statistics(self) -> Dict[str, float]:
+        """Compute comprehensive latency statistics."""
+        if not self.results:
+            return {}
+
+        latencies = self.get_latencies()
+
+        return {
+            "count": len(latencies),
+            "min_ms": min(latencies),
+            "max_ms": max(latencies),
+            "mean_ms": statistics.mean(latencies),
+            "median_ms": statistics.median(latencies),
+            "p50_ms": np.percentile(latencies, 50),
+            "p90_ms": np.percentile(latencies, 90),
+            "p95_ms": np.percentile(latencies, 95),  # The main metric
+            "p99_ms": np.percentile(latencies, 99),
+            "std_dev_ms": statistics.stdev(latencies) if len(latencies) > 1 else 0,
+        }
+
+    def print_statistics(self):
+        """Print formatted statistics to console."""
+        stats = self.compute_statistics()
+        if not stats:
+            print("No results to analyze.")
+            return
+
+        print("\n" + "=" * 50)
+        print("SEARCH LATENCY ANALYSIS")
+        print("=" * 50)
+        print(f"Total searches: {stats['count']}")
+        print(f"Min latency: {stats['min_ms']:.2f} ms")
+        print(f"Max latency: {stats['max_ms']:.2f} ms")
+        print(f"Mean latency: {stats['mean_ms']:.2f} ms")
+        print(f"Median latency: {stats['median_ms']:.2f} ms")
+        print("-" * 30)
+        print("PERCENTILES:")
+        print(f"P50 (median): {stats['p50_ms']:.2f} ms")
+        print(f"P90: {stats['p90_ms']:.2f} ms")
+        print(f"P95: {stats['p95_ms']:.2f} ms ⭐")  # Highlighted
+        print(f"P99: {stats['p99_ms']:.2f} ms")
+        print("-" * 30)
+        print(f"Standard deviation: {stats['std_dev_ms']:.2f} ms")
+        print("=" * 50)
+
+
+with open("./fixture/memobase/results_0710_3000.json", "r") as f:
+    data = json.load(f)
+latencies = []
+
+
+for k in data.keys():
+    for d in data[k]:
+        latencies.append(d["speaker_1_memory_time"] * 1000)
+        latencies.append(d["speaker_2_memory_time"] * 1000)
+
+
+analyzer = LatencyAnalyzer(latencies)
+analyzer.print_statistics()