Upload cuda_training_transformers_image-classification_google/vit-base-patch16-224/benchmark.json with huggingface_hub

cuda_training_transformers_image-classification_google/vit-base-patch16-224/benchmark.json

@@ -3,7 +3,7 @@
         "name": "cuda_training_transformers_image-classification_google/vit-base-patch16-224",
         "backend": {
             "name": "pytorch",
-            "version": "2.4.0+cu124",
+            "version": "2.4.0+cu121",
             "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
             "task": "image-classification",
             "library": "transformers",
@@ -110,7 +110,7 @@
         "overall": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1611.599872,
+                "max_ram": 1504.268288,
                 "max_global_vram": 2611.478528,
                 "max_process_vram": 0.0,
                 "max_reserved": 1956.642816,
@@ -119,24 +119,24 @@
             "latency": {
                 "unit": "s",
                 "count": 5,
-                "total": 0.4841891784667969,
-                "mean": 0.09683783569335938,
-                "stdev": 0.11274135330278746,
-                "p50": 0.04044595336914063,
-                "p90": 0.20967996826171875,
-                "p95": 0.2660001708984374,
-                "p99": 0.3110563330078125,
+                "total": 0.4983961639404297,
+                "mean": 0.09967923278808594,
+                "stdev": 0.11825831340910486,
+                "p50": 0.04063846588134765,
+                "p90": 0.21801738433837892,
+                "p95": 0.277106485748291,
+                "p99": 0.3243777668762207,
                 "values": [
-                    0.32232037353515625,
-                    0.0407193603515625,
-                    0.04039884948730469,
-                    0.04044595336914063,
-                    0.040304641723632816
+                    0.33619558715820314,
+                    0.04063846588134765,
+                    0.040750080108642575,
+                    0.04021657562255859,
+                    0.04059545516967773
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 103.26542232589102
+                "value": 100.3217994390025
             },
             "energy": null,
             "efficiency": null
@@ -144,7 +144,7 @@
         "warmup": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1611.599872,
+                "max_ram": 1504.268288,
                 "max_global_vram": 2611.478528,
                 "max_process_vram": 0.0,
                 "max_reserved": 1956.642816,
@@ -153,21 +153,21 @@
             "latency": {
                 "unit": "s",
                 "count": 2,
-                "total": 0.36303973388671873,
-                "mean": 0.18151986694335936,
-                "stdev": 0.14080050659179688,
-                "p50": 0.18151986694335936,
-                "p90": 0.29416027221679686,
-                "p95": 0.3082403228759765,
-                "p99": 0.3195043634033203,
+                "total": 0.3768340530395508,
+                "mean": 0.1884170265197754,
+                "stdev": 0.14777856063842776,
+                "p50": 0.1884170265197754,
+                "p90": 0.3066398750305176,
+                "p95": 0.32141773109436034,
+                "p99": 0.33324001594543456,
                 "values": [
-                    0.32232037353515625,
-                    0.0407193603515625
+                    0.33619558715820314,
+                    0.04063846588134765
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 22.036155421203244
+                "value": 21.229503903566687
             },
             "energy": null,
             "efficiency": null
@@ -175,7 +175,7 @@
         "train": {
             "memory": {
                 "unit": "MB",
-                "max_ram": 1611.599872,
+                "max_ram": 1504.268288,
                 "max_global_vram": 2611.478528,
                 "max_process_vram": 0.0,
                 "max_reserved": 1956.642816,
@@ -184,22 +184,22 @@
             "latency": {
                 "unit": "s",
                 "count": 3,
-                "total": 0.12114944458007815,
-                "mean": 0.040383148193359385,
-                "stdev": 5.874886247238727e-05,
-                "p50": 0.04039884948730469,
-                "p90": 0.04043653259277344,
-                "p95": 0.04044124298095703,
-                "p99": 0.04044501129150391,
+                "total": 0.1215621109008789,
+                "mean": 0.0405207036336263,
+                "stdev": 0.00022412437510523794,
+                "p50": 0.04059545516967773,
+                "p90": 0.04071915512084961,
+                "p95": 0.04073461761474609,
+                "p99": 0.040746987609863276,
                 "values": [
-                    0.04039884948730469,
-                    0.04044595336914063,
-                    0.040304641723632816
+                    0.040750080108642575,
+                    0.04021657562255859,
+                    0.04059545516967773
                 ]
             },
             "throughput": {
                 "unit": "samples/s",
-                "value": 148.57682643441458
+                "value": 148.07245338703524
             },
             "energy": null,
             "efficiency": null