From 94bf9ae4e9b8199636668ccbe4dabcdc3b9e5ae6 Mon Sep 17 00:00:00 2001
From: Andy Dai <76841985+Imss27@users.noreply.github.com>
Date: Thu, 10 Oct 2024 17:33:16 -0700
Subject: [PATCH] [Misc] Fix sampling from sonnet for long context case (#9235)

---
 benchmarks/benchmark_serving.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/benchmarks/benchmark_serving.py b/benchmarks/benchmark_serving.py
index 292d1f37fbf3e..04999518b7138 100644
--- a/benchmarks/benchmark_serving.py
+++ b/benchmarks/benchmark_serving.py
@@ -176,9 +176,9 @@ def sample_sonnet_requests(
     # Sample the rest of lines per request.
     sampled_requests: List[Tuple[str, int, int]] = []
     for _ in range(num_requests):
-        sampled_lines = "".join(
-            prefix_lines +
-            random.sample(poem_lines, num_input_lines - num_prefix_lines))
+        num_lines_needed = num_input_lines - num_prefix_lines
+        sampled_lines = "".join(prefix_lines +
+                                random.choices(poem_lines, k=num_lines_needed))
 
         prompt = f"{base_prompt}{sampled_lines}"
         message = [
@@ -536,7 +536,7 @@ def process_one_metric(
         # E.g., "Time to First Token"
         metric_header: str,
     ):
-        # This function print and add statistics of the specified
+        # This function prints and adds statistics of the specified
         # metric.
         if metric_attribute_name not in selected_percentile_metrics:
             return