Results from GH actions on NVIDIA_RTX4090x1

Author: arjunsuresh

closed/GATEOverflow/compliance/RTX4090x1-nvidia-gpu-TensorRT-default_config/retinanet/multistream/TEST01/accuracy/accuracy.txt

@@ -1,2 +1,2 @@
 
-hash=19aa9cd9d9559e9f36724c8d4c1cf02116fa611f2a5ccaf16f3168ff391514a5
+hash=aba591d1b53637089186ed5c7721bfcd739fa7849f9e5b61636111b8a3167dc7

closed/GATEOverflow/compliance/RTX4090x1-nvidia-gpu-TensorRT-default_config/retinanet/multistream/TEST01/accuracy/mlperf_log_accuracy.json

@@ -4,35 +4,35 @@ MLPerf Results Summary
 SUT name : LWIS_Server
 Scenario : MultiStream
 Mode     : PerformanceOnly
-99th percentile latency (ns) : 10868648
+99th percentile latency (ns) : 10929584
 Result is : VALID
   Min duration satisfied : Yes
   Min queries satisfied : Yes
   Early stopping satisfied: Yes
 Early Stopping Result:
- * Processed at least 662 queries (55813).
- * Would discard 502 highest latency queries.
- * Early stopping 99th percentile estimate: 10877944
+ * Processed at least 662 queries (55658).
+ * Would discard 501 highest latency queries.
+ * Early stopping 99th percentile estimate: 10938382
 
 ================================================
 Additional Stats
 ================================================
 Per-query latency:  
-Min latency (ns)                : 10524679
-Max latency (ns)                : 11688525
-Mean latency (ns)               : 10713508
-50.00 percentile latency (ns)   : 10708143
-90.00 percentile latency (ns)   : 10804043
-95.00 percentile latency (ns)   : 10822075
-97.00 percentile latency (ns)   : 10832565
-99.00 percentile latency (ns)   : 10868648
-99.90 percentile latency (ns)   : 11003735
+Min latency (ns)                : 10537951
+Max latency (ns)                : 11715990
+Mean latency (ns)               : 10741941
+50.00 percentile latency (ns)   : 10737306
+90.00 percentile latency (ns)   : 10837306
+95.00 percentile latency (ns)   : 10855030
+97.00 percentile latency (ns)   : 10865565
+99.00 percentile latency (ns)   : 10929584
+99.90 percentile latency (ns)   : 11060952
 
 ================================================
 Test Parameters Used
 ================================================
 samples_per_query : 8
-target_qps : 233.027
+target_qps : 232.532
 target_latency (ns): 0
 max_async_queries : 1
 min_duration (ms): 600000

closed/GATEOverflow/compliance/RTX4090x1-nvidia-gpu-TensorRT-default_config/retinanet/multistream/TEST01/performance/run_1/mlperf_log_detail.txt

@@ -1,4 +1,4 @@
 Verifying performance.
-reference score = 10906536
-test score = 10877944
+reference score = 10908244
+test score = 10938382
 TEST PASS

closed/GATEOverflow/compliance/RTX4090x1-nvidia-gpu-TensorRT-default_config/retinanet/multistream/TEST01/performance/run_1/mlperf_log_summary.txt

@@ -1,2 +1,2 @@
 
-hash=3f1d2ff9b7ca089dd7250ed41fb1b4b79dfd7447df8c3b33e3261b09fbb25253
+hash=ead1da660a46db00ae1e10b3c8e40aef022da5f18167a948618f844cd080607a

closed/GATEOverflow/compliance/RTX4090x1-nvidia-gpu-TensorRT-default_config/retinanet/multistream/TEST01/verify_performance.txt

@@ -4,7 +4,7 @@ MLPerf Results Summary
 SUT name : LWIS_Server
 Scenario : Offline
 Mode     : PerformanceOnly
-Samples per second: 867.49
+Samples per second: 863.646
 Result is : VALID
   Min duration satisfied : Yes
   Min queries satisfied : Yes
@@ -13,21 +13,21 @@ Result is : VALID
 ================================================
 Additional Stats
 ================================================
-Min latency (ns)                : 5708744
-Max latency (ns)                : 665015001528
-Mean latency (ns)               : 331698376937
-50.00 percentile latency (ns)   : 331434934944
-90.00 percentile latency (ns)   : 598334667318
-95.00 percentile latency (ns)   : 631683215249
-97.00 percentile latency (ns)   : 645018895640
-99.00 percentile latency (ns)   : 658357074711
-99.90 percentile latency (ns)   : 664352177107
+Min latency (ns)                : 5654542
+Max latency (ns)                : 665356225508
+Mean latency (ns)               : 332138673725
+50.00 percentile latency (ns)   : 332038841442
+90.00 percentile latency (ns)   : 598679380830
+95.00 percentile latency (ns)   : 632013956415
+97.00 percentile latency (ns)   : 645356649048
+99.00 percentile latency (ns)   : 658694375820
+99.90 percentile latency (ns)   : 664692100968
 
 ================================================
 Test Parameters Used
 ================================================
-samples_per_query : 576894
-target_qps : 874.083
+samples_per_query : 574632
+target_qps : 870.655
 target_latency (ns): 0
 max_async_queries : 1
 min_duration (ms): 600000

closed/GATEOverflow/compliance/RTX4090x1-nvidia-gpu-TensorRT-default_config/retinanet/offline/TEST01/accuracy/accuracy.txt

@@ -4,8 +4,8 @@ Reading performance mode results...
 num_acc_log_entries = 24781
 num_acc_log_duplicate_keys = 0
 num_acc_log_data_mismatch = 0
-num_perf_log_entries = 255
-num_perf_log_qsl_idx_match = 255
+num_perf_log_entries = 257
+num_perf_log_qsl_idx_match = 257
 num_perf_log_data_mismatch = 0
 num_missing_qsl_idxs = 0
 TEST PASS