Hugging Face's logo

Datasets:
optimum-benchmark
/
cuda

Dataset card Data Studio Files Community
IlyasMoutawwakil HF staff commited on
Commit
833e37b
·
verified ·
1 Parent(s): 8023b51

Upload cuda_training_transformers_multiple-choice_FacebookAI/roberta-base/benchmark_report.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. cuda_training_transformers_multiple-choice_FacebookAI/roberta-base/benchmark_report.json +57 -54
cuda_training_transformers_multiple-choice_FacebookAI/roberta-base/benchmark_report.json CHANGED
@@ -2,73 +2,75 @@
2
  "overall": {
3
  "memory": {
4
  "unit": "MB",
5
- "max_ram": 1333.239808,
6
- "max_global_vram": 3384.27904,
7
  "max_process_vram": 0.0,
8
- "max_reserved": 2728.394752,
9
  "max_allocated": 2516.23424
10
  },
11
  "latency": {
12
  "unit": "s",
13
- "count": 5,
14
- "total": 0.5310904350280762,
15
- "mean": 0.10621808700561523,
16
- "stdev": 0.12178600287446865,
17
- "p50": 0.04538675308227539,
18
- "p90": 0.2282371078491211,
19
- "p95": 0.28901314620971674,
20
- "p99": 0.33763397689819336,
21
  "values": [
22
- 0.3497891845703125,
23
- 0.045908992767333984,
24
- 0.04498739242553711,
25
- 0.04538675308227539,
26
- 0.045018112182617184
27
- ]
 
 
 
 
 
 
 
 
 
28
  },
29
  "throughput": {
30
  "unit": "samples/s",
31
- "value": 94.14592450221166
32
  },
33
  "energy": {
34
  "unit": "kWh",
35
- "cpu": 1.0077815970796817e-05,
36
- "ram": 5.504013079623437e-06,
37
- "gpu": 1.6129457347968668e-05,
38
- "total": 3.171128639838892e-05
39
  },
40
  "efficiency": {
41
  "unit": "samples/kWh",
42
- "value": 315345.1384585914
43
  }
44
  },
45
  "warmup": {
46
  "memory": {
47
  "unit": "MB",
48
- "max_ram": 1333.239808,
49
- "max_global_vram": 3384.27904,
50
  "max_process_vram": 0.0,
51
- "max_reserved": 2728.394752,
52
  "max_allocated": 2516.23424
53
  },
54
  "latency": {
55
  "unit": "s",
56
- "count": 2,
57
- "total": 0.39569817733764645,
58
- "mean": 0.19784908866882323,
59
- "stdev": 0.15194009590148924,
60
- "p50": 0.19784908866882323,
61
- "p90": 0.3194011653900146,
62
- "p95": 0.33459517498016356,
63
- "p99": 0.3467503826522827,
64
  "values": [
65
- 0.3497891845703125,
66
- 0.045908992767333984
67
- ]
 
 
 
 
 
 
 
 
 
68
  },
69
  "throughput": {
70
  "unit": "samples/s",
71
- "value": 20.21742949089618
72
  },
73
  "energy": null,
74
  "efficiency": null
@@ -76,31 +78,32 @@
76
  "train": {
77
  "memory": {
78
  "unit": "MB",
79
- "max_ram": 1333.239808,
80
- "max_global_vram": 3384.27904,
81
  "max_process_vram": 0.0,
82
- "max_reserved": 2728.394752,
83
  "max_allocated": 2516.23424
84
  },
85
  "latency": {
86
  "unit": "s",
87
- "count": 3,
88
- "total": 0.13539225769042967,
89
- "mean": 0.04513075256347656,
90
- "stdev": 0.00018145362141645116,
91
- "p50": 0.045018112182617184,
92
- "p90": 0.04531302490234375,
93
- "p95": 0.045349888992309566,
94
- "p99": 0.045379380264282225,
95
  "values": [
96
- 0.04498739242553711,
97
- 0.04538675308227539,
98
- 0.045018112182617184
99
- ]
 
 
 
 
 
 
 
 
 
100
  },
101
  "throughput": {
102
  "unit": "samples/s",
103
- "value": 132.94704074701568
104
  },
105
  "energy": null,
106
  "efficiency": null
 
2
  "overall": {
3
  "memory": {
4
  "unit": "MB",
5
+ "max_ram": 1315.258368,
6
+ "max_global_vram": 3386.376192,
7
  "max_process_vram": 0.0,
8
+ "max_reserved": 2730.491904,
9
  "max_allocated": 2516.23424
10
  },
11
  "latency": {
12
  "unit": "s",
 
 
 
 
 
 
 
 
13
  "values": [
14
+ 0.3432120361328125,
15
+ 0.05156966400146484,
16
+ 0.05033574295043945,
17
+ 0.05009408187866211,
18
+ 0.0501319694519043
19
+ ],
20
+ "count": 5,
21
+ "total": 0.5453434944152832,
22
+ "mean": 0.10906869888305663,
23
+ "p50": 0.05033574295043945,
24
+ "p90": 0.22655508728027346,
25
+ "p95": 0.2848835617065429,
26
+ "p99": 0.33154634124755855,
27
+ "stdev": 0.11707292174555654,
28
+ "stdev_": 107.33869840244637
29
  },
30
  "throughput": {
31
  "unit": "samples/s",
32
+ "value": 91.68533321115338
33
  },
34
  "energy": {
35
  "unit": "kWh",
36
+ "cpu": 1.0429607820373752e-05,
37
+ "ram": 5.693287145294507e-06,
38
+ "gpu": 1.8337792448086532e-05,
39
+ "total": 3.446068741375479e-05
40
  },
41
  "efficiency": {
42
  "unit": "samples/kWh",
43
+ "value": 290185.7377345455
44
  }
45
  },
46
  "warmup": {
47
  "memory": {
48
  "unit": "MB",
49
+ "max_ram": 1315.258368,
50
+ "max_global_vram": 3386.376192,
51
  "max_process_vram": 0.0,
52
+ "max_reserved": 2730.491904,
53
  "max_allocated": 2516.23424
54
  },
55
  "latency": {
56
  "unit": "s",
 
 
 
 
 
 
 
 
57
  "values": [
58
+ 0.3432120361328125,
59
+ 0.05156966400146484
60
+ ],
61
+ "count": 2,
62
+ "total": 0.3947817001342773,
63
+ "mean": 0.19739085006713866,
64
+ "p50": 0.19739085006713866,
65
+ "p90": 0.3140477989196777,
66
+ "p95": 0.3286299175262451,
67
+ "p99": 0.340295612411499,
68
+ "stdev": 0.14582118606567382,
69
+ "stdev_": 73.87433916824187
70
  },
71
  "throughput": {
72
  "unit": "samples/s",
73
+ "value": 20.264363817469135
74
  },
75
  "energy": null,
76
  "efficiency": null
 
78
  "train": {
79
  "memory": {
80
  "unit": "MB",
81
+ "max_ram": 1315.258368,
82
+ "max_global_vram": 3386.376192,
83
  "max_process_vram": 0.0,
84
+ "max_reserved": 2730.491904,
85
  "max_allocated": 2516.23424
86
  },
87
  "latency": {
88
  "unit": "s",
 
 
 
 
 
 
 
 
89
  "values": [
90
+ 0.05033574295043945,
91
+ 0.05009408187866211,
92
+ 0.0501319694519043
93
+ ],
94
+ "count": 3,
95
+ "total": 0.15056179428100586,
96
+ "mean": 0.05018726476033528,
97
+ "p50": 0.0501319694519043,
98
+ "p90": 0.05029498825073242,
99
+ "p95": 0.050315365600585936,
100
+ "p99": 0.05033166748046875,
101
+ "stdev": 0.00010612318865043078,
102
+ "stdev_": 0.21145441808238088
103
  },
104
  "throughput": {
105
  "unit": "samples/s",
106
+ "value": 119.5522415627242
107
  },
108
  "energy": null,
109
  "efficiency": null