Auto-merge updates from auto-update branch

Author: mlcommons-bot

open/MLCommons/measurements/RTX4090x2-nvidia_original-gpu-tensorrt-vdefault-default_config/bert-99.9/offline/README.md

@@ -19,7 +19,7 @@ pip install -U cmind
 
 cm rm cache -f
 
-cm pull repo mlcommons@mlperf-automations --checkout=467517e4a572872046058e394a0d83512cfff38b
+cm pull repo mlcommons@mlperf-automations --checkout=c52956b27fa8d06ec8db53f885e1f05021e379e9
 
 cm run script \
 	--tags=app,mlperf,inference,generic,_nvidia,_bert-99.9,_tensorrt,_cuda,_valid,_r4.1-dev_default,_offline \
@@ -71,7 +71,7 @@ cm run script \
 	--env.CM_DOCKER_REUSE_EXISTING_CONTAINER=yes \
 	--env.CM_DOCKER_DETACHED_MODE=yes \
 	--env.CM_MLPERF_INFERENCE_RESULTS_DIR_=/home/arjun/gh_action_results/valid_results \
-	--env.CM_DOCKER_CONTAINER_ID=d7e43b0ba70a \
+	--env.CM_DOCKER_CONTAINER_ID=0b3b13aa449e \
 	--env.CM_MLPERF_LOADGEN_COMPLIANCE_TEST=TEST01 \
 	--add_deps_recursive.compiler.tags=gcc \
 	--add_deps_recursive.coco2014-original.tags=_full \
@@ -129,4 +129,4 @@ Model Precision: fp16
 `F1`: `90.88324`, Required accuracy for closed division `>= 90.78313`
 
 ### Performance Results 
-`Samples per second`: `3360.15`
+`Samples per second`: `3348.97`

open/MLCommons/measurements/RTX4090x2-nvidia_original-gpu-tensorrt-vdefault-default_config/bert-99.9/offline/accuracy_console.out

@@ -1,7 +1,7 @@
-[2024-12-27 21:28:18,054 main.py:229 INFO] Detected system ID: KnownSystem.RTX4090x2
-[2024-12-27 21:28:18,579 generate_conf_files.py:107 INFO] Generated measurements/ entries for RTX4090x2_TRT/bert-99.9/Offline
-[2024-12-27 21:28:18,579 __init__.py:46 INFO] Running command: ./build/bin/harness_bert --logfile_outdir="/cm-mount/home/arjun/gh_action_results/valid_results/RTX4090x2-nvidia_original-gpu-tensorrt-vdefault-default_config/bert-99.9/offline/accuracy" --logfile_prefix="mlperf_log_" --performance_sample_count=10833 --test_mode="AccuracyOnly" --gpu_batch_size=256 --mlperf_conf_path="/home/cmuser/CM/repos/local/cache/13da9a9a9e4e460f/inference/mlperf.conf" --tensor_path="build/preprocessed_data/squad_tokenized/input_ids.npy,build/preprocessed_data/squad_tokenized/segment_ids.npy,build/preprocessed_data/squad_tokenized/input_mask.npy" --use_graphs=false --user_conf_path="/home/cmuser/CM/repos/mlcommons@mlperf-automations/script/generate-mlperf-inference-user-conf/tmp/1e993d10f7d444a99c5cc2482bfdc85c.conf" --gpu_inference_streams=2 --gpu_copy_streams=2 --gpu_engines="./build/engines/RTX4090x2/bert/Offline/bert-Offline-gpu-fp16_S_384_B_256_P_2_vs.custom_k_99_9_MaxP.plan" --scenario Offline --model bert
-[2024-12-27 21:28:18,579 __init__.py:53 INFO] Overriding Environment
+[2024-12-28 21:43:12,494 main.py:229 INFO] Detected system ID: KnownSystem.RTX4090x2
+[2024-12-28 21:43:13,048 generate_conf_files.py:107 INFO] Generated measurements/ entries for RTX4090x2_TRT/bert-99.9/Offline
+[2024-12-28 21:43:13,048 __init__.py:46 INFO] Running command: ./build/bin/harness_bert --logfile_outdir="/cm-mount/home/arjun/gh_action_results/valid_results/RTX4090x2-nvidia_original-gpu-tensorrt-vdefault-default_config/bert-99.9/offline/accuracy" --logfile_prefix="mlperf_log_" --performance_sample_count=10833 --test_mode="AccuracyOnly" --gpu_batch_size=256 --mlperf_conf_path="/home/cmuser/CM/repos/local/cache/11daf5a55e5449f4/inference/mlperf.conf" --tensor_path="build/preprocessed_data/squad_tokenized/input_ids.npy,build/preprocessed_data/squad_tokenized/segment_ids.npy,build/preprocessed_data/squad_tokenized/input_mask.npy" --use_graphs=false --user_conf_path="/home/cmuser/CM/repos/mlcommons@mlperf-automations/script/generate-mlperf-inference-user-conf/tmp/c6913ddf30654865abebcb8dbfb0abd9.conf" --gpu_inference_streams=2 --gpu_copy_streams=2 --gpu_engines="./build/engines/RTX4090x2/bert/Offline/bert-Offline-gpu-fp16_S_384_B_256_P_2_vs.custom_k_99_9_MaxP.plan" --scenario Offline --model bert
+[2024-12-28 21:43:13,048 __init__.py:53 INFO] Overriding Environment
 benchmark : Benchmark.BERT
 buffer_manager_thread_count : 0
 coalesced_tensor : True
@@ -11,8 +11,8 @@ gpu_copy_streams : 2
 gpu_inference_streams : 2
 input_dtype : int32
 input_format : linear
-log_dir : /home/cmuser/CM/repos/local/cache/94a57f78972843c6/repo/closed/NVIDIA/build/logs/2024.12.27-21.28.16
-mlperf_conf_path : /home/cmuser/CM/repos/local/cache/13da9a9a9e4e460f/inference/mlperf.conf
+log_dir : /home/cmuser/CM/repos/local/cache/94a57f78972843c6/repo/closed/NVIDIA/build/logs/2024.12.28-21.43.11
+mlperf_conf_path : /home/cmuser/CM/repos/local/cache/11daf5a55e5449f4/inference/mlperf.conf
 offline_expected_qps : 0.0
 precision : fp16
 preprocessed_data_dir : /home/cmuser/CM/repos/local/cache/4db00c74da1e44c8/preprocessed_data
@@ -21,7 +21,7 @@ system : SystemConfiguration(host_cpu_conf=CPUConfiguration(layout={CPU(name='In
 tensor_path : build/preprocessed_data/squad_tokenized/input_ids.npy,build/preprocessed_data/squad_tokenized/segment_ids.npy,build/preprocessed_data/squad_tokenized/input_mask.npy
 test_mode : AccuracyOnly
 use_graphs : False
-user_conf_path : /home/cmuser/CM/repos/mlcommons@mlperf-automations/script/generate-mlperf-inference-user-conf/tmp/1e993d10f7d444a99c5cc2482bfdc85c.conf
+user_conf_path : /home/cmuser/CM/repos/mlcommons@mlperf-automations/script/generate-mlperf-inference-user-conf/tmp/c6913ddf30654865abebcb8dbfb0abd9.conf
 system_id : RTX4090x2
 config_name : RTX4090x2_bert_Offline
 workload_setting : WorkloadSetting(HarnessType.Custom, AccuracyTarget.k_99_9, PowerSetting.MaxP)
@@ -34,64 +34,64 @@ skip_file_checks : True
 power_limit : None
 cpu_freq : None
 &&&& RUNNING BERT_HARNESS # ./build/bin/harness_bert
-I1227 21:28:18.626047 20263 main_bert.cc:163] Found 2 GPUs
-I1227 21:28:18.748579 20263 bert_server.cc:147] Engine Path: ./build/engines/RTX4090x2/bert/Offline/bert-Offline-gpu-fp16_S_384_B_256_P_2_vs.custom_k_99_9_MaxP.plan
-[I] [TRT] Loaded engine size: 699 MiB
-[I] [TRT] Loaded engine size: 699 MiB
+I1228 21:43:13.096694 20264 main_bert.cc:163] Found 2 GPUs
+I1228 21:43:13.218209 20264 bert_server.cc:147] Engine Path: ./build/engines/RTX4090x2/bert/Offline/bert-Offline-gpu-fp16_S_384_B_256_P_2_vs.custom_k_99_9_MaxP.plan
+[I] [TRT] Loaded engine size: 700 MiB
+[I] [TRT] Loaded engine size: 700 MiB
 [W] [TRT] Using an engine plan file across different models of devices is not recommended and is likely to affect performance or even cause errors.
 [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +7, GPU +10, now: CPU 1008, GPU 1511 (MiB)
-[I] [TRT] [MemUsageChange] Init cuDNN: CPU +1, GPU +10, now: CPU 1009, GPU 1521 (MiB)
+[I] [TRT] [MemUsageChange] Init cuDNN: CPU +2, GPU +10, now: CPU 1010, GPU 1521 (MiB)
 [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +1152, now: CPU 0, GPU 1152 (MiB)
-[I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +6, GPU +10, now: CPU 1017, GPU 1256 (MiB)
-[I] [TRT] [MemUsageChange] Init cuDNN: CPU +1, GPU +10, now: CPU 1018, GPU 1266 (MiB)
+[I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +6, GPU +10, now: CPU 1018, GPU 1255 (MiB)
+[I] [TRT] [MemUsageChange] Init cuDNN: CPU +1, GPU +10, now: CPU 1019, GPU 1265 (MiB)
 [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +1, GPU +576, now: CPU 1, GPU 1152 (MiB)
-I1227 21:28:19.399394 20263 bert_server.cc:208] Engines Creation Completed
-I1227 21:28:19.432389 20263 bert_core_vs.cc:385] Engine - Device Memory requirements: 1409287680
-I1227 21:28:19.432394 20263 bert_core_vs.cc:393] Engine - Number of Optimization Profiles: 2
-I1227 21:28:19.432399 20263 bert_core_vs.cc:415] Engine - Profile 0 maxDims 98304 Bmax=256 Smax=384
+I1228 21:43:13.877281 20264 bert_server.cc:208] Engines Creation Completed
+I1228 21:43:13.902434 20264 bert_core_vs.cc:385] Engine - Device Memory requirements: 1409287680
+I1228 21:43:13.902443 20264 bert_core_vs.cc:393] Engine - Number of Optimization Profiles: 2
+I1228 21:43:13.902449 20264 bert_core_vs.cc:415] Engine - Profile 0 maxDims 98304 Bmax=256 Smax=384
 [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +8, now: CPU 319, GPU 2859 (MiB)
-[I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +8, now: CPU 319, GPU 2867 (MiB)
-I1227 21:28:19.498804 20263 bert_core_vs.cc:426] Setting Opt.Prof. to 0
-I1227 21:28:19.498829 20263 bert_core_vs.cc:444] Context creation complete. Max supported batchSize: 256
+[I] [TRT] [MemUsageChange] Init cuDNN: CPU +1, GPU +8, now: CPU 320, GPU 2867 (MiB)
+I1228 21:43:13.967170 20264 bert_core_vs.cc:426] Setting Opt.Prof. to 0
+I1228 21:43:13.967195 20264 bert_core_vs.cc:444] Context creation complete. Max supported batchSize: 256
 [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +0, now: CPU 1, GPU 1152 (MiB)
-I1227 21:28:19.499617 20263 bert_core_vs.cc:476] Setup complete
-I1227 21:28:19.499763 20263 bert_core_vs.cc:385] Engine - Device Memory requirements: 1409287680
-I1227 21:28:19.499768 20263 bert_core_vs.cc:393] Engine - Number of Optimization Profiles: 2
-I1227 21:28:19.499770 20263 bert_core_vs.cc:415] Engine - Profile 0 maxDims 98304 Bmax=256 Smax=384
-[I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +8, now: CPU 442, GPU 2604 (MiB)
-[I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +8, now: CPU 442, GPU 2612 (MiB)
-I1227 21:28:19.565945 20263 bert_core_vs.cc:426] Setting Opt.Prof. to 0
-I1227 21:28:19.565958 20263 bert_core_vs.cc:444] Context creation complete. Max supported batchSize: 256
+I1228 21:43:13.967973 20264 bert_core_vs.cc:476] Setup complete
+I1228 21:43:13.968132 20264 bert_core_vs.cc:385] Engine - Device Memory requirements: 1409287680
+I1228 21:43:13.968134 20264 bert_core_vs.cc:393] Engine - Number of Optimization Profiles: 2
+I1228 21:43:13.968137 20264 bert_core_vs.cc:415] Engine - Profile 0 maxDims 98304 Bmax=256 Smax=384
+[I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +8, now: CPU 443, GPU 2603 (MiB)
+[I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +8, now: CPU 443, GPU 2611 (MiB)
+I1228 21:43:14.032351 20264 bert_core_vs.cc:426] Setting Opt.Prof. to 0
+I1228 21:43:14.032366 20264 bert_core_vs.cc:444] Context creation complete. Max supported batchSize: 256
 [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +1, GPU +0, now: CPU 2, GPU 1152 (MiB)
-I1227 21:28:19.566730 20263 bert_core_vs.cc:476] Setup complete
-I1227 21:28:19.566908 20263 bert_core_vs.cc:385] Engine - Device Memory requirements: 1409287680
-I1227 21:28:19.566911 20263 bert_core_vs.cc:393] Engine - Number of Optimization Profiles: 2
-I1227 21:28:19.566915 20263 bert_core_vs.cc:415] Engine - Profile 1 maxDims 98304 Bmax=256 Smax=384
+I1228 21:43:14.033135 20264 bert_core_vs.cc:476] Setup complete
+I1228 21:43:14.033304 20264 bert_core_vs.cc:385] Engine - Device Memory requirements: 1409287680
+I1228 21:43:14.033309 20264 bert_core_vs.cc:393] Engine - Number of Optimization Profiles: 2
+I1228 21:43:14.033313 20264 bert_core_vs.cc:415] Engine - Profile 1 maxDims 98304 Bmax=256 Smax=384
 [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +8, now: CPU 566, GPU 4345 (MiB)
 [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +10, now: CPU 566, GPU 4355 (MiB)
-I1227 21:28:19.632443 20263 bert_core_vs.cc:426] Setting Opt.Prof. to 1
+I1228 21:43:14.097095 20264 bert_core_vs.cc:426] Setting Opt.Prof. to 1
 [I] [TRT] Could not set default profile 0 for execution context. Profile index must be set explicitly.
 [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +1, GPU +0, now: CPU 3, GPU 1152 (MiB)
-I1227 21:28:19.632763 20263 bert_core_vs.cc:444] Context creation complete. Max supported batchSize: 256
-I1227 21:28:19.633572 20263 bert_core_vs.cc:476] Setup complete
-I1227 21:28:19.633745 20263 bert_core_vs.cc:385] Engine - Device Memory requirements: 1409287680
-I1227 21:28:19.633749 20263 bert_core_vs.cc:393] Engine - Number of Optimization Profiles: 2
-I1227 21:28:19.633752 20263 bert_core_vs.cc:415] Engine - Profile 1 maxDims 98304 Bmax=256 Smax=384
-[I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +8, now: CPU 689, GPU 4090 (MiB)
-[I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +10, now: CPU 689, GPU 4100 (MiB)
-I1227 21:28:19.699026 20263 bert_core_vs.cc:426] Setting Opt.Prof. to 1
+I1228 21:43:14.097440 20264 bert_core_vs.cc:444] Context creation complete. Max supported batchSize: 256
+I1228 21:43:14.098248 20264 bert_core_vs.cc:476] Setup complete
+I1228 21:43:14.098415 20264 bert_core_vs.cc:385] Engine - Device Memory requirements: 1409287680
+I1228 21:43:14.098419 20264 bert_core_vs.cc:393] Engine - Number of Optimization Profiles: 2
+I1228 21:43:14.098423 20264 bert_core_vs.cc:415] Engine - Profile 1 maxDims 98304 Bmax=256 Smax=384
+[I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +8, now: CPU 690, GPU 4089 (MiB)
+[I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +10, now: CPU 690, GPU 4099 (MiB)
+I1228 21:43:14.162550 20264 bert_core_vs.cc:426] Setting Opt.Prof. to 1
 [I] [TRT] Could not set default profile 0 for execution context. Profile index must be set explicitly.
 [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +0, now: CPU 3, GPU 1152 (MiB)
-I1227 21:28:19.699366 20263 bert_core_vs.cc:444] Context creation complete. Max supported batchSize: 256
-I1227 21:28:19.700189 20263 bert_core_vs.cc:476] Setup complete
-I1227 21:28:20.895964 20263 main_bert.cc:184] Starting running actual test.
-I1227 21:28:24.215916 20263 main_bert.cc:190] Finished running actual test.
+I1228 21:43:14.162904 20264 bert_core_vs.cc:444] Context creation complete. Max supported batchSize: 256
+I1228 21:43:14.163702 20264 bert_core_vs.cc:476] Setup complete
+I1228 21:43:15.365103 20264 main_bert.cc:184] Starting running actual test.
+I1228 21:43:18.679199 20264 main_bert.cc:190] Finished running actual test.
 
 No warnings encountered during test.
 
 No errors encountered during test.
-[2024-12-27 21:28:24,446 run_harness.py:166 INFO] Result: Accuracy run detected.
-[2024-12-27 21:28:24,446 __init__.py:46 INFO] Running command: PYTHONPATH=code/bert/tensorrt/helpers python3 /home/cmuser/CM/repos/local/cache/94a57f78972843c6/repo/closed/NVIDIA/build/inference/language/bert/accuracy-squad.py --log_file /cm-mount/home/arjun/gh_action_results/valid_results/RTX4090x2-nvidia_original-gpu-tensorrt-vdefault-default_config/bert-99.9/offline/accuracy/mlperf_log_accuracy.json --vocab_file build/models/bert/vocab.txt --val_data /home/cmuser/CM/repos/local/cache/4db00c74da1e44c8/data/squad/dev-v1.1.json --out_file /cm-mount/home/arjun/gh_action_results/valid_results/RTX4090x2-nvidia_original-gpu-tensorrt-vdefault-default_config/bert-99.9/offline/accuracy/predictions.json --output_dtype float16
+[2024-12-28 21:43:18,912 run_harness.py:166 INFO] Result: Accuracy run detected.
+[2024-12-28 21:43:18,912 __init__.py:46 INFO] Running command: PYTHONPATH=code/bert/tensorrt/helpers python3 /home/cmuser/CM/repos/local/cache/94a57f78972843c6/repo/closed/NVIDIA/build/inference/language/bert/accuracy-squad.py --log_file /cm-mount/home/arjun/gh_action_results/valid_results/RTX4090x2-nvidia_original-gpu-tensorrt-vdefault-default_config/bert-99.9/offline/accuracy/mlperf_log_accuracy.json --vocab_file build/models/bert/vocab.txt --val_data /home/cmuser/CM/repos/local/cache/4db00c74da1e44c8/data/squad/dev-v1.1.json --out_file /cm-mount/home/arjun/gh_action_results/valid_results/RTX4090x2-nvidia_original-gpu-tensorrt-vdefault-default_config/bert-99.9/offline/accuracy/predictions.json --output_dtype float16
 {"exact_match": 83.67076631977294, "f1": 90.8832407068292}
 Reading examples...
 Loading cached features from 'eval_features.pickle'...