Upload cuda_training_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub

cuda_training_transformers_text-generation_openai-community/gpt2/benchmark.json

@@ -3,7 +3,7 @@
         "name": "cuda_training_transformers_text-generation_openai-community/gpt2",
         "backend": {
             "name": "pytorch",
-            "version": "2.4.0+rocm6.1",
+            "version": "2.3.1+rocm5.7",
             "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
             "task": "text-generation",
             "library": "transformers",
@@ -71,7 +71,7 @@
             "name": "process",
             "_target_": "optimum_benchmark.launchers.process.launcher.ProcessLauncher",
             "device_isolation": true,
-            "device_isolation_action": "error",
+            "device_isolation_action": "warn",
             "numactl": false,
             "numactl_kwargs": {},
             "start_method": "spawn"
@@ -101,7 +101,7 @@
             "optimum_benchmark_commit": null,
             "transformers_version": "4.44.2",
             "transformers_commit": null,
-            "accelerate_version": "0.33.0",
+            "accelerate_version": "0.34.0",
             "accelerate_commit": null,
             "diffusers_version": "0.30.2",
             "diffusers_commit": null,
@@ -117,33 +117,33 @@
         "overall": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1639.968768,
-                "max_global_vram": 11.268096,
-                "max_process_vram": 0.0,
-                "max_reserved": 2889.875456,
+                "max_ram": 1293.139968,
+                "max_global_vram": 68702.69952,
+                "max_process_vram": 346401.763328,
+                "max_reserved": 2894.06976,
                 "max_allocated": 2506.73664
             },
             "latency": {
                 "unit": "s",
                 "count": 5,
-                "total": 0.7898254585266112,
-                "mean": 0.15796509170532225,
-                "stdev": 0.21940817402775237,
-                "p50": 0.048201904296875,
-                "p90": 0.37763274688720705,
-                "p95": 0.48720681533813465,
-                "p99": 0.5748660700988769,
+                "total": 0.7399985160827637,
+                "mean": 0.14799970321655273,
+                "stdev": 0.21410563119152728,
+                "p50": 0.04097893142700195,
+                "p90": 0.3622956146240235,
+                "p95": 0.469253123474121,
+                "p99": 0.5548191305541992,
                 "values": [
-                    0.5967808837890625,
-                    0.048910541534423826,
-                    0.04789342498779297,
-                    0.048038703918457035,
-                    0.048201904296875
+                    0.5762106323242188,
+                    0.04142308807373047,
+                    0.04097893142700195,
+                    0.04069317245483398,
+                    0.04069269180297851
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 63.305125784718385
+                "value": 67.56770306064756
             },
             "energy": null,
             "efficiency": null
@@ -151,30 +151,30 @@
         "warmup": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1639.968768,
-                "max_global_vram": 11.268096,
-                "max_process_vram": 0.0,
-                "max_reserved": 2889.875456,
+                "max_ram": 1293.139968,
+                "max_global_vram": 68702.69952,
+                "max_process_vram": 346401.763328,
+                "max_reserved": 2894.06976,
                 "max_allocated": 2506.73664
             },
             "latency": {
                 "unit": "s",
                 "count": 2,
-                "total": 0.6456914253234863,
-                "mean": 0.32284571266174317,
-                "stdev": 0.27393517112731935,
-                "p50": 0.32284571266174317,
-                "p90": 0.5419938495635986,
-                "p95": 0.5693873666763305,
-                "p99": 0.591302180366516,
+                "total": 0.6176337203979493,
+                "mean": 0.30881686019897464,
+                "stdev": 0.2673937721252442,
+                "p50": 0.30881686019897464,
+                "p90": 0.52273187789917,
+                "p95": 0.5494712551116944,
+                "p99": 0.5708627568817138,
                 "values": [
-                    0.5967808837890625,
-                    0.048910541534423826
+                    0.5762106323242188,
+                    0.04142308807373047
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 12.389819170963998
+                "value": 12.952660672162617
             },
             "energy": null,
             "efficiency": null
@@ -182,31 +182,31 @@
         "train": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1639.968768,
-                "max_global_vram": 11.268096,
-                "max_process_vram": 0.0,
-                "max_reserved": 2889.875456,
+                "max_ram": 1293.139968,
+                "max_global_vram": 68702.69952,
+                "max_process_vram": 346401.763328,
+                "max_reserved": 2894.06976,
                 "max_allocated": 2506.73664
             },
             "latency": {
                 "unit": "s",
                 "count": 3,
-                "total": 0.144134033203125,
-                "mean": 0.048044677734375006,
-                "stdev": 0.0001260069730620708,
-                "p50": 0.048038703918457035,
-                "p90": 0.04816926422119141,
-                "p95": 0.0481855842590332,
-                "p99": 0.04819864028930664,
+                "total": 0.12236479568481445,
+                "mean": 0.04078826522827148,
+                "stdev": 0.00013482150486302612,
+                "p50": 0.04069317245483398,
+                "p90": 0.04092177963256836,
+                "p95": 0.040950355529785154,
+                "p99": 0.04097321624755859,
                 "values": [
-                    0.04789342498779297,
-                    0.048038703918457035,
-                    0.048201904296875
+                    0.04097893142700195,
+                    0.04069317245483398,
+                    0.04069269180297851
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 124.88375992804548
+                "value": 147.10113230903562
             },
             "energy": null,
             "efficiency": null