Upload cuda_training_transformers_fill-mask_hf-internal-testing/tiny-random-BertModel/benchmark_report.json with huggingface_hub
Browse files
cuda_training_transformers_fill-mask_hf-internal-testing/tiny-random-BertModel/benchmark_report.json
ADDED
@@ -0,0 +1,111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"overall": {
|
3 |
+
"memory": {
|
4 |
+
"unit": "MB",
|
5 |
+
"max_ram": 1292.378112,
|
6 |
+
"max_global_vram": 681.050112,
|
7 |
+
"max_process_vram": 0.0,
|
8 |
+
"max_reserved": 25.165824,
|
9 |
+
"max_allocated": 19.652608
|
10 |
+
},
|
11 |
+
"latency": {
|
12 |
+
"unit": "s",
|
13 |
+
"values": [
|
14 |
+
0.29976882934570315,
|
15 |
+
0.01428889560699463,
|
16 |
+
0.01345952033996582,
|
17 |
+
0.013425663948059082,
|
18 |
+
0.013402112007141113
|
19 |
+
],
|
20 |
+
"count": 5,
|
21 |
+
"total": 0.3543450212478638,
|
22 |
+
"mean": 0.07086900424957276,
|
23 |
+
"p50": 0.01345952033996582,
|
24 |
+
"p90": 0.18557685585021977,
|
25 |
+
"p95": 0.2426728425979614,
|
26 |
+
"p99": 0.2883496319961548,
|
27 |
+
"stdev": 0.11445039843950845,
|
28 |
+
"stdev_": 161.495705564677
|
29 |
+
},
|
30 |
+
"throughput": {
|
31 |
+
"unit": "samples/s",
|
32 |
+
"value": 141.10541139796365
|
33 |
+
},
|
34 |
+
"energy": {
|
35 |
+
"unit": "kWh",
|
36 |
+
"cpu": 6.4100997583308435e-06,
|
37 |
+
"ram": 3.4907546266820437e-06,
|
38 |
+
"gpu": 8.121950941998579e-06,
|
39 |
+
"total": 1.8022805327011465e-05
|
40 |
+
},
|
41 |
+
"efficiency": {
|
42 |
+
"unit": "samples/kWh",
|
43 |
+
"value": 554852.578084091
|
44 |
+
}
|
45 |
+
},
|
46 |
+
"warmup": {
|
47 |
+
"memory": {
|
48 |
+
"unit": "MB",
|
49 |
+
"max_ram": 1292.378112,
|
50 |
+
"max_global_vram": 681.050112,
|
51 |
+
"max_process_vram": 0.0,
|
52 |
+
"max_reserved": 25.165824,
|
53 |
+
"max_allocated": 19.652608
|
54 |
+
},
|
55 |
+
"latency": {
|
56 |
+
"unit": "s",
|
57 |
+
"values": [
|
58 |
+
0.29976882934570315,
|
59 |
+
0.01428889560699463
|
60 |
+
],
|
61 |
+
"count": 2,
|
62 |
+
"total": 0.3140577249526978,
|
63 |
+
"mean": 0.1570288624763489,
|
64 |
+
"p50": 0.1570288624763489,
|
65 |
+
"p90": 0.27122083597183233,
|
66 |
+
"p95": 0.2854948326587677,
|
67 |
+
"p99": 0.29691403000831607,
|
68 |
+
"stdev": 0.14273996686935428,
|
69 |
+
"stdev_": 90.90046544204779
|
70 |
+
},
|
71 |
+
"throughput": {
|
72 |
+
"unit": "samples/s",
|
73 |
+
"value": 25.47302411110865
|
74 |
+
},
|
75 |
+
"energy": null,
|
76 |
+
"efficiency": null
|
77 |
+
},
|
78 |
+
"train": {
|
79 |
+
"memory": {
|
80 |
+
"unit": "MB",
|
81 |
+
"max_ram": 1292.378112,
|
82 |
+
"max_global_vram": 681.050112,
|
83 |
+
"max_process_vram": 0.0,
|
84 |
+
"max_reserved": 25.165824,
|
85 |
+
"max_allocated": 19.652608
|
86 |
+
},
|
87 |
+
"latency": {
|
88 |
+
"unit": "s",
|
89 |
+
"values": [
|
90 |
+
0.01345952033996582,
|
91 |
+
0.013425663948059082,
|
92 |
+
0.013402112007141113
|
93 |
+
],
|
94 |
+
"count": 3,
|
95 |
+
"total": 0.04028729629516602,
|
96 |
+
"mean": 0.013429098765055339,
|
97 |
+
"p50": 0.013425663948059082,
|
98 |
+
"p90": 0.013452749061584472,
|
99 |
+
"p95": 0.013456134700775146,
|
100 |
+
"p99": 0.013458843212127685,
|
101 |
+
"stdev": 2.3562366113167512e-05,
|
102 |
+
"stdev_": 0.1754575383307222
|
103 |
+
},
|
104 |
+
"throughput": {
|
105 |
+
"unit": "samples/s",
|
106 |
+
"value": 446.79096527407773
|
107 |
+
},
|
108 |
+
"energy": null,
|
109 |
+
"efficiency": null
|
110 |
+
}
|
111 |
+
}
|