Upload cuda_training_transformers_fill-mask_google-bert/bert-base-uncased/benchmark.json with huggingface_hub

cuda_training_transformers_fill-mask_google-bert/bert-base-uncased/benchmark.json

@@ -3,7 +3,7 @@
         "name": "cuda_training_transformers_fill-mask_google-bert/bert-base-uncased",
         "backend": {
             "name": "pytorch",
-            "version": "2.4.0+cu121",
+            "version": "2.4.0+cu124",
             "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
             "task": "fill-mask",
             "library": "transformers",
@@ -110,7 +110,7 @@
         "overall": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1154.29376,
+                "max_ram": 1282.228224,
                 "max_global_vram": 3169.32096,
                 "max_process_vram": 0.0,
                 "max_reserved": 2520.776704,
@@ -119,24 +119,24 @@
             "latency": {
                 "unit": "s",
                 "count": 5,
-                "total": 0.8239503135681153,
-                "mean": 0.16479006271362306,
-                "stdev": 0.24471952980441253,
-                "p50": 0.0425082893371582,
-                "p90": 0.40976443939208995,
-                "p95": 0.5319964469909667,
-                "p99": 0.6297820530700684,
+                "total": 0.5063363265991211,
+                "mean": 0.10126726531982422,
+                "stdev": 0.11545870066599251,
+                "p50": 0.04336640167236328,
+                "p90": 0.21699301605224613,
+                "p95": 0.2745883201599121,
+                "p99": 0.32066456344604494,
                 "values": [
-                    0.6542284545898438,
-                    0.043068416595458986,
-                    0.0425082893371582,
-                    0.04180070495605469,
-                    0.04234444808959961
+                    0.33218362426757814,
+                    0.04420710372924805,
+                    0.04331417465209961,
+                    0.04326502227783203,
+                    0.04336640167236328
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 60.683270795146726
+                "value": 98.74859332300332
             },
             "energy": null,
             "efficiency": null
@@ -144,7 +144,7 @@
         "warmup": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1154.29376,
+                "max_ram": 1282.228224,
                 "max_global_vram": 3169.32096,
                 "max_process_vram": 0.0,
                 "max_reserved": 2520.776704,
@@ -153,21 +153,21 @@
             "latency": {
                 "unit": "s",
                 "count": 2,
-                "total": 0.6972968711853028,
-                "mean": 0.3486484355926514,
-                "stdev": 0.30558001899719245,
-                "p50": 0.3486484355926514,
-                "p90": 0.5931124507904053,
-                "p95": 0.6236704526901246,
-                "p99": 0.6481168542099,
+                "total": 0.37639072799682616,
+                "mean": 0.18819536399841308,
+                "stdev": 0.14398826026916503,
+                "p50": 0.18819536399841308,
+                "p90": 0.30338597221374514,
+                "p95": 0.31778479824066164,
+                "p99": 0.32930385906219484,
                 "values": [
-                    0.6542284545898438,
-                    0.043068416595458986
+                    0.33218362426757814,
+                    0.04420710372924805
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 11.472875228023279
+                "value": 21.25450869254
             },
             "energy": null,
             "efficiency": null
@@ -175,7 +175,7 @@
         "train": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1154.29376,
+                "max_ram": 1282.228224,
                 "max_global_vram": 3169.32096,
                 "max_process_vram": 0.0,
                 "max_reserved": 2520.776704,
@@ -184,22 +184,22 @@
             "latency": {
                 "unit": "s",
                 "count": 3,
-                "total": 0.1266534423828125,
-                "mean": 0.04221781412760417,
-                "stdev": 0.0003024301950791384,
-                "p50": 0.04234444808959961,
-                "p90": 0.04247552108764648,
-                "p95": 0.04249190521240234,
-                "p99": 0.04250501251220703,
+                "total": 0.1299455986022949,
+                "mean": 0.0433151995340983,
+                "stdev": 4.1394308757077004e-05,
+                "p50": 0.04331417465209961,
+                "p90": 0.04335595626831055,
+                "p95": 0.043361178970336915,
+                "p99": 0.04336535713195801,
                 "values": [
-                    0.0425082893371582,
-                    0.04180070495605469,
-                    0.04234444808959961
+                    0.04331417465209961,
+                    0.04326502227783203,
+                    0.04336640167236328
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 142.1201008149044
+                "value": 138.519505036026
             },
             "energy": null,
             "efficiency": null