Results from self hosted Github actions - NVIDIARTX4090

GATEOverflow · Oct 10, 2024 · b905763 · b905763
1 parent 7554d5b
commit b905763
Show file tree

Hide file tree

Showing 8 changed files with 210 additions and 209 deletions.
diff --git a/open/MLCommons/measurements/97c3ec750d0e-nvidia-gpu-TensorRT-scc24-base/README.md b/open/MLCommons/measurements/97c3ec750d0e-nvidia-gpu-TensorRT-scc24-base/README.md
@@ -1,3 +1,3 @@
 | Model               | Scenario   | Accuracy              |   Throughput | Latency (in ms)   |
 |---------------------|------------|-----------------------|--------------|-------------------|
-| stable-diffusion-xl | offline    | (15.47816, 232.17873) |        1.134 | -                 |
+| stable-diffusion-xl | offline    | (15.47816, 232.17873) |        1.141 | -                 |
diff --git a/...c3ec750d0e-nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/README.md b/...c3ec750d0e-nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/README.md
@@ -19,7 +19,7 @@ pip install -U cmind
 
 cm rm cache -f
 
-cm pull repo gateoverflow@cm4mlops --checkout=50431589d01cd60060a0d1be3c1f8193d99c6144
+cm pull repo gateoverflow@cm4mlops --checkout=52f2ebb1436d6e255d60a3dc0acf99e4f24f492d
 
 cm run script \
 	--tags=app,mlperf,inference,generic,_nvidia,_sdxl,_tensorrt,_test,_r4.1-dev_default,_float16,_offline \
@@ -93,4 +93,4 @@ Model Precision: int8
 `FID_SCORE`: `232.17873`, Required accuracy for closed division `>= 23.01086` and `<= 23.95008`
 
 ### Performance Results 
-`Samples per second`: `1.13424`
+`Samples per second`: `1.14052`
diff --git a/...3ec750d0e-nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/cm-version-info.json b/...3ec750d0e-nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/cm-version-info.json
diff --git a/...s/97c3ec750d0e-nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/pip_freeze.json b/...s/97c3ec750d0e-nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/pip_freeze.json
@@ -38,6 +38,7 @@
     "humanfriendly": "10.0",
     "hypothesis": "6.113.0",
     "idna": "3.10",
+    "ijson": "3.3.0",
     "importlib-metadata": "8.5.0",
     "jinja2": "3.1.4",
     "jmespath": "1.0.1",
@@ -51,7 +52,7 @@
     "multiprocess": "0.70.17",
     "networkx": "3.1",
     "ninja": "1.11.1.1",
-    "numpy": "1.23.5",
+    "numpy": "1.22.0",
     "nvidia-ammo": "0.7.4",
     "nvidia-dali-cuda120": "1.42.0",
     "nvidia-nvimgcodec-cu12": "0.3.0.5",

diff --git a/...ec750d0e-nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/accuracy/accuracy.txt b/...ec750d0e-nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/accuracy/accuracy.txt
@@ -1,2 +1,2 @@
-Accuracy Results: {'FID_SCORE': 232.1787311146868, 'CLIP_SCORE': 15.478164181113243}
-hash=2859af71e624d4ee21d591e4b9c1fbbeac2ab06549a62151f79963a348eabd7d
+Accuracy Results: {'FID_SCORE': 232.17873111420926, 'CLIP_SCORE': 15.478164181113243}
+hash=be4a6a67ec0aff031aff864996583fee58e2be816daf6654e832d531493bbc82
diff --git a/...nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/accuracy/mlperf_log_detail.txt b/...nvidia-gpu-TensorRT-scc24-base/stable-diffusion-xl/offline/accuracy/mlperf_log_detail.txt
diff --git a/...u-TensorRT-scc24-base/stable-diffusion-xl/offline/performance/run_1/mlperf_log_detail.txt b/...u-TensorRT-scc24-base/stable-diffusion-xl/offline/performance/run_1/mlperf_log_detail.txt
diff --git a/...-TensorRT-scc24-base/stable-diffusion-xl/offline/performance/run_1/mlperf_log_summary.txt b/...-TensorRT-scc24-base/stable-diffusion-xl/offline/performance/run_1/mlperf_log_summary.txt
@@ -4,7 +4,7 @@ MLPerf Results Summary
 SUT name : PySUT
 Scenario : Offline
 Mode     : PerformanceOnly
-Samples per second: 1.13424
+Samples per second: 1.14052
 Result is : VALID
   Min duration satisfied : Yes
   Min queries satisfied : Yes
@@ -13,15 +13,15 @@ Result is : VALID
 ================================================
 Additional Stats
 ================================================
-Min latency (ns)                : 2941389477
-Max latency (ns)                : 8816469794
-Mean latency (ns)               : 5293948242
-50.00 percentile latency (ns)   : 5881108255
-90.00 percentile latency (ns)   : 8816469794
-95.00 percentile latency (ns)   : 8816469794
-97.00 percentile latency (ns)   : 8816469794
-99.00 percentile latency (ns)   : 8816469794
-99.90 percentile latency (ns)   : 8816469794
+Min latency (ns)                : 2927810275
+Max latency (ns)                : 8767960151
+Mean latency (ns)               : 5267954137
+50.00 percentile latency (ns)   : 5846589423
+90.00 percentile latency (ns)   : 8767960151
+95.00 percentile latency (ns)   : 8767960151
+97.00 percentile latency (ns)   : 8767960151
+99.00 percentile latency (ns)   : 8767960151
+99.90 percentile latency (ns)   : 8767960151
 
 ================================================
 Test Parameters Used