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+cu124",
+            "version": "2.4.0+cu121",
             "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
             "task": "fill-mask",
             "library": "transformers",
@@ -110,7 +110,7 @@
         "overall": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1282.056192,
+                "max_ram": 1154.838528,
                 "max_global_vram": 3169.32096,
                 "max_process_vram": 0.0,
                 "max_reserved": 2520.776704,
@@ -119,24 +119,24 @@
             "latency": {
                 "unit": "s",
                 "count": 5,
-                "total": 0.5029416427612304,
-                "mean": 0.10058832855224609,
-                "stdev": 0.11439831019396637,
-                "p50": 0.04344518280029297,
-                "p90": 0.21540864562988282,
-                "p95": 0.2723952713012695,
-                "p99": 0.3179845718383789,
+                "total": 0.8190555839538574,
+                "mean": 0.16381111679077148,
+                "stdev": 0.24238888085378063,
+                "p50": 0.04297420883178711,
+                "p90": 0.40638486328125,
+                "p95": 0.5274865905761718,
+                "p99": 0.6243679724121093,
                 "values": [
-                    0.32938189697265624,
-                    0.04444876861572265,
-                    0.04281139373779297,
-                    0.04344518280029297,
-                    0.04285440063476562
+                    0.6485883178710937,
+                    0.043079681396484375,
+                    0.042183616638183596,
+                    0.04297420883178711,
+                    0.042229759216308595
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 99.41511250786864
+                "value": 61.045917981088834
             },
             "energy": null,
             "efficiency": null
@@ -144,7 +144,7 @@
         "warmup": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1282.056192,
+                "max_ram": 1154.838528,
                 "max_global_vram": 3169.32096,
                 "max_process_vram": 0.0,
                 "max_reserved": 2520.776704,
@@ -153,21 +153,21 @@
             "latency": {
                 "unit": "s",
                 "count": 2,
-                "total": 0.3738306655883789,
-                "mean": 0.18691533279418945,
-                "stdev": 0.1424665641784668,
-                "p50": 0.18691533279418945,
-                "p90": 0.3008885841369629,
-                "p95": 0.31513524055480957,
-                "p99": 0.3265325656890869,
+                "total": 0.691667999267578,
+                "mean": 0.345833999633789,
+                "stdev": 0.3027543182373047,
+                "p50": 0.345833999633789,
+                "p90": 0.5880374542236328,
+                "p95": 0.6183128860473632,
+                "p99": 0.6425332315063477,
                 "values": [
-                    0.32938189697265624,
-                    0.04444876861572265
+                    0.6485883178710937,
+                    0.043079681396484375
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 21.40006354858196
+                "value": 11.566242776117111
             },
             "energy": null,
             "efficiency": null
@@ -175,7 +175,7 @@
         "train": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1282.056192,
+                "max_ram": 1154.838528,
                 "max_global_vram": 3169.32096,
                 "max_process_vram": 0.0,
                 "max_reserved": 2520.776704,
@@ -184,22 +184,22 @@
             "latency": {
                 "unit": "s",
                 "count": 3,
-                "total": 0.12911097717285155,
-                "mean": 0.04303699239095052,
-                "stdev": 0.0002891677206584111,
-                "p50": 0.04285440063476562,
-                "p90": 0.0433270263671875,
-                "p95": 0.04338610458374023,
-                "p99": 0.04343336715698242,
+                "total": 0.1273875846862793,
+                "mean": 0.04246252822875977,
+                "stdev": 0.0003623028788319013,
+                "p50": 0.042229759216308595,
+                "p90": 0.04282531890869141,
+                "p95": 0.042899763870239256,
+                "p99": 0.04295931983947754,
                 "values": [
-                    0.04281139373779297,
-                    0.04344518280029297,
-                    0.04285440063476562
+                    0.042183616638183596,
+                    0.04297420883178711,
+                    0.042229759216308595
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 139.41494669273482
+                "value": 141.30105413591966
             },
             "energy": null,
             "efficiency": null