File size: 6,170 Bytes
69def8e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 | import argparse
import json
import os
import time
import sys
import numpy as np
import csv
from datetime import datetime, timezone
from concurrent.futures import ThreadPoolExecutor, as_completed
# Add workspace to path
sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
from scripts.inference import run_inference, load_inference_config, DEFAULT_CONFIG_PATH
def load_test(
pairs_csv,
concurrency=4,
num_requests=20,
threshold=0.40,
model_name='Facenet',
confidence_k=10.0,
score_is_distance=False,
output_path='outputs/runtime/load_test_summary.json',
):
"""
Simulate concurrent inference requests.
"""
pairs = []
with open(pairs_csv, 'r') as f:
reader = csv.DictReader(f)
for row in reader:
pairs.append((row['left_path'], row['right_path']))
if not pairs:
print("Error: No pairs found for load test.")
return
# Limit requests if needed
test_pairs = pairs[:num_requests]
if len(test_pairs) < num_requests:
# Recycle pairs if needed
test_pairs = (test_pairs * (num_requests // len(test_pairs) + 1))[:num_requests]
print(f"Starting Load Test: {num_requests} requests with concurrency={concurrency}")
latencies = []
success_count = 0
failure_count = 0
failure_examples = []
start_time = time.perf_counter()
with ThreadPoolExecutor(max_workers=concurrency) as executor:
futures = [
executor.submit(
run_inference,
p[0],
p[1],
threshold,
model_name,
confidence_k,
score_is_distance,
)
for p in test_pairs
]
for future in as_completed(futures):
try:
res = future.result()
latencies.append(res['latency_total_ms'])
success_count += 1
except Exception as e:
print(f"Request failed: {e}")
failure_count += 1
if len(failure_examples) < 5:
failure_examples.append(str(e))
total_time = time.perf_counter() - start_time
# Calculate stats
if latencies:
avg_latency = float(np.mean(latencies))
p50_latency = float(np.percentile(latencies, 50))
p95_latency = float(np.percentile(latencies, 95))
p99_latency = float(np.percentile(latencies, 99))
min_latency = float(np.min(latencies))
max_latency = float(np.max(latencies))
throughput = success_count / total_time
else:
avg_latency = p50_latency = p95_latency = p99_latency = 0.0
min_latency = max_latency = 0.0
throughput = 0.0
summary = {
'timestamp_utc': datetime.now(timezone.utc).isoformat(),
'pairs_csv': pairs_csv,
'concurrency': int(concurrency),
'total_requests': int(num_requests),
'success_count': int(success_count),
'failure_count': int(failure_count),
'failure_examples': failure_examples,
'total_time_s': round(float(total_time), 4),
'throughput_rps': round(float(throughput), 4),
'threshold': float(threshold),
'model_name': str(model_name),
'confidence_k': float(confidence_k),
'score_is_distance': bool(score_is_distance),
'latency_ms': {
'avg': round(avg_latency, 4),
'p50': round(p50_latency, 4),
'p95': round(p95_latency, 4),
'p99': round(p99_latency, 4),
'min': round(min_latency, 4),
'max': round(max_latency, 4),
},
}
if output_path:
os.makedirs(os.path.dirname(output_path), exist_ok=True)
with open(output_path, 'w') as f:
json.dump(summary, f, indent=2)
print("\n--- Load Test Results ---")
print(f"Total Requests: {num_requests}")
print(f"Success Count: {success_count}")
print(f"Failure Count: {failure_count}")
print(f"Total Time: {total_time:.2f}s")
print(f"Throughput: {throughput:.2f} req/s")
print(f"Avg Latency: {avg_latency:.2f} ms")
print(f"p95 Latency: {p95_latency:.2f} ms")
print(f"p99 Latency: {p99_latency:.2f} ms")
if output_path:
print(f"Summary JSON: {output_path}")
print("-------------------------\n")
return summary
def main():
parser = argparse.ArgumentParser(description="FaceID Load Test CLI (Milestone 3)")
parser.add_argument("--pairs", type=str, default="outputs/pairs_v2/val.csv", help="Pairs CSV for test data")
parser.add_argument("--concurrency", type=int, default=4, help="Number of concurrent workers")
parser.add_argument("--requests", type=int, default=20, help="Total number of requests to perform")
parser.add_argument(
"--config",
type=str,
default=DEFAULT_CONFIG_PATH,
help="Inference config used for model/threshold settings.",
)
parser.add_argument(
"--threshold",
type=float,
default=None,
help="Optional threshold override. If omitted, uses config threshold.",
)
parser.add_argument(
"--output",
type=str,
default="outputs/runtime/load_test_summary.json",
help="Path for JSON summary output.",
)
args = parser.parse_args()
if not os.path.exists(args.pairs):
print(f"Error: Pairs file {args.pairs} not found. Run make_pairs.py or recalibrate first.")
return
runtime_config = load_inference_config(args.config)
threshold = args.threshold if args.threshold is not None else runtime_config['threshold']
load_test(
args.pairs,
concurrency=args.concurrency,
num_requests=args.requests,
threshold=threshold,
model_name=runtime_config['model_name'],
confidence_k=runtime_config['confidence_k'],
score_is_distance=runtime_config['score_is_distance'],
output_path=args.output,
)
if __name__ == "__main__":
main()
|