From 95d3671ed1ec6919eb52a748d7f3de128f3d5897 Mon Sep 17 00:00:00 2001
From: Bo <a@a.com>
Date: Wed, 3 Jul 2024 02:07:54 +0800
Subject: [PATCH] accurate benchmark

---
 RWKV_v6_demo_cuda_bf16.py | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/RWKV_v6_demo_cuda_bf16.py b/RWKV_v6_demo_cuda_bf16.py
index 75c3d5c..09e7d5d 100644
--- a/RWKV_v6_demo_cuda_bf16.py
+++ b/RWKV_v6_demo_cuda_bf16.py
@@ -270,10 +270,10 @@ def decode(self, tokens):
     min_time = 1e10
     min_time_all = 1e10
 
-    t000 = time.time()
+    t000 = time.perf_counter()
 
     for i in range(LENGTH_PER_TRIAL):
-        t00 = time.time()
+        t00 = time.perf_counter()
         token = sample_logits(out, TEMPERATURE, TOP_P)
         all_tokens += [token]
         try:
@@ -283,14 +283,14 @@ def decode(self, tokens):
                 out_last = i + 1
         except:
             pass
-        t0 = time.time()
+        t0 = time.perf_counter()
 
         out, state = model.forward(token, state)
         
-        t1 = time.time()
+        t1 = time.perf_counter()
         min_time = min(min_time, t1 - t0)
         min_time_all = min(min_time_all, t1 - t00)
     
-    print(f'\n[ {round(1/min_time_all,2)} (real) / {round(1/min_time,2)} (ignore sampling & tokenizer) token/s = {round(time.time()-t000,3)}s ]', end='')
+    print(f'\n[ {round(1/min_time_all,2)} (real) / {round(1/min_time,2)} (ignore sampling & tokenizer) token/s = {round(time.perf_counter()-t000,3)}s ]', end='')
 
 print('\n')