Commit
•
817363a
1
Parent(s):
7ec8c77
Upload cuda_training_transformers_fill-mask_google-bert/bert-base-uncased/benchmark.json with huggingface_hub
Browse files
cuda_training_transformers_fill-mask_google-bert/bert-base-uncased/benchmark.json
CHANGED
@@ -6,19 +6,17 @@
|
|
6 |
"version": "2.2.2+rocm5.7",
|
7 |
"_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
|
8 |
"task": "fill-mask",
|
9 |
-
"model": "google-bert/bert-base-uncased",
|
10 |
"library": "transformers",
|
|
|
|
|
11 |
"device": "cuda",
|
12 |
"device_ids": "0",
|
13 |
"seed": 42,
|
14 |
"inter_op_num_threads": null,
|
15 |
"intra_op_num_threads": null,
|
16 |
-
"
|
17 |
-
|
18 |
-
|
19 |
-
"local_files_only": false,
|
20 |
-
"trust_remote_code": false
|
21 |
-
},
|
22 |
"no_weights": true,
|
23 |
"device_map": null,
|
24 |
"torch_dtype": null,
|
@@ -88,7 +86,7 @@
|
|
88 |
"gpu_count": 1,
|
89 |
"gpu_vram_mb": 68702699520,
|
90 |
"optimum_benchmark_version": "0.2.0",
|
91 |
-
"optimum_benchmark_commit": "
|
92 |
"transformers_version": "4.40.2",
|
93 |
"transformers_commit": null,
|
94 |
"accelerate_version": "0.30.1",
|
@@ -107,33 +105,33 @@
|
|
107 |
"overall": {
|
108 |
"memory": {
|
109 |
"unit": "MB",
|
110 |
-
"max_ram":
|
111 |
-
"max_global_vram": 2945.
|
112 |
-
"max_process_vram":
|
113 |
"max_reserved": 2497.708032,
|
114 |
"max_allocated": 2195.345408
|
115 |
},
|
116 |
"latency": {
|
117 |
"unit": "s",
|
118 |
"count": 5,
|
119 |
-
"total": 0.
|
120 |
-
"mean": 0.
|
121 |
-
"stdev": 0.
|
122 |
-
"p50": 0.
|
123 |
-
"p90": 0.
|
124 |
-
"p95": 0.
|
125 |
-
"p99": 0.
|
126 |
"values": [
|
127 |
-
0.
|
128 |
-
0.
|
129 |
-
0.
|
130 |
-
0.
|
131 |
-
0.
|
132 |
]
|
133 |
},
|
134 |
"throughput": {
|
135 |
"unit": "samples/s",
|
136 |
-
"value":
|
137 |
},
|
138 |
"energy": null,
|
139 |
"efficiency": null
|
@@ -141,30 +139,30 @@
|
|
141 |
"warmup": {
|
142 |
"memory": {
|
143 |
"unit": "MB",
|
144 |
-
"max_ram":
|
145 |
-
"max_global_vram": 2945.
|
146 |
-
"max_process_vram":
|
147 |
"max_reserved": 2497.708032,
|
148 |
"max_allocated": 2195.345408
|
149 |
},
|
150 |
"latency": {
|
151 |
"unit": "s",
|
152 |
"count": 2,
|
153 |
-
"total": 0.
|
154 |
-
"mean": 0.
|
155 |
-
"stdev": 0.
|
156 |
-
"p50": 0.
|
157 |
-
"p90": 0.
|
158 |
-
"p95": 0.
|
159 |
-
"p99": 0.
|
160 |
"values": [
|
161 |
-
0.
|
162 |
-
0.
|
163 |
]
|
164 |
},
|
165 |
"throughput": {
|
166 |
"unit": "samples/s",
|
167 |
-
"value": 13.
|
168 |
},
|
169 |
"energy": null,
|
170 |
"efficiency": null
|
@@ -172,31 +170,31 @@
|
|
172 |
"train": {
|
173 |
"memory": {
|
174 |
"unit": "MB",
|
175 |
-
"max_ram":
|
176 |
-
"max_global_vram": 2945.
|
177 |
-
"max_process_vram":
|
178 |
"max_reserved": 2497.708032,
|
179 |
"max_allocated": 2195.345408
|
180 |
},
|
181 |
"latency": {
|
182 |
"unit": "s",
|
183 |
"count": 3,
|
184 |
-
"total": 0.
|
185 |
-
"mean": 0.
|
186 |
-
"stdev": 0.
|
187 |
-
"p50": 0.
|
188 |
-
"p90": 0.
|
189 |
-
"p95": 0.
|
190 |
-
"p99": 0.
|
191 |
"values": [
|
192 |
-
0.
|
193 |
-
0.
|
194 |
-
0.
|
195 |
]
|
196 |
},
|
197 |
"throughput": {
|
198 |
"unit": "samples/s",
|
199 |
-
"value":
|
200 |
},
|
201 |
"energy": null,
|
202 |
"efficiency": null
|
|
|
6 |
"version": "2.2.2+rocm5.7",
|
7 |
"_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
|
8 |
"task": "fill-mask",
|
|
|
9 |
"library": "transformers",
|
10 |
+
"model": "google-bert/bert-base-uncased",
|
11 |
+
"processor": "google-bert/bert-base-uncased",
|
12 |
"device": "cuda",
|
13 |
"device_ids": "0",
|
14 |
"seed": 42,
|
15 |
"inter_op_num_threads": null,
|
16 |
"intra_op_num_threads": null,
|
17 |
+
"model_kwargs": {},
|
18 |
+
"processor_kwargs": {},
|
19 |
+
"hub_kwargs": {},
|
|
|
|
|
|
|
20 |
"no_weights": true,
|
21 |
"device_map": null,
|
22 |
"torch_dtype": null,
|
|
|
86 |
"gpu_count": 1,
|
87 |
"gpu_vram_mb": 68702699520,
|
88 |
"optimum_benchmark_version": "0.2.0",
|
89 |
+
"optimum_benchmark_commit": "6fd377459e287bb09e9383ba2516b1b2a271a562",
|
90 |
"transformers_version": "4.40.2",
|
91 |
"transformers_commit": null,
|
92 |
"accelerate_version": "0.30.1",
|
|
|
105 |
"overall": {
|
106 |
"memory": {
|
107 |
"unit": "MB",
|
108 |
+
"max_ram": 1116.532736,
|
109 |
+
"max_global_vram": 2945.032192,
|
110 |
+
"max_process_vram": 296709.005312,
|
111 |
"max_reserved": 2497.708032,
|
112 |
"max_allocated": 2195.345408
|
113 |
},
|
114 |
"latency": {
|
115 |
"unit": "s",
|
116 |
"count": 5,
|
117 |
+
"total": 0.7349412956237793,
|
118 |
+
"mean": 0.14698825912475585,
|
119 |
+
"stdev": 0.21148885124944541,
|
120 |
+
"p50": 0.04117127990722656,
|
121 |
+
"p90": 0.35872324066162115,
|
122 |
+
"p95": 0.4643443473815917,
|
123 |
+
"p99": 0.5488412327575684,
|
124 |
"values": [
|
125 |
+
0.5699654541015625,
|
126 |
+
0.04185992050170898,
|
127 |
+
0.04117127990722656,
|
128 |
+
0.04092536163330078,
|
129 |
+
0.04101927947998047
|
130 |
]
|
131 |
},
|
132 |
"throughput": {
|
133 |
"unit": "samples/s",
|
134 |
+
"value": 68.03264464485241
|
135 |
},
|
136 |
"energy": null,
|
137 |
"efficiency": null
|
|
|
139 |
"warmup": {
|
140 |
"memory": {
|
141 |
"unit": "MB",
|
142 |
+
"max_ram": 1116.532736,
|
143 |
+
"max_global_vram": 2945.032192,
|
144 |
+
"max_process_vram": 296709.005312,
|
145 |
"max_reserved": 2497.708032,
|
146 |
"max_allocated": 2195.345408
|
147 |
},
|
148 |
"latency": {
|
149 |
"unit": "s",
|
150 |
"count": 2,
|
151 |
+
"total": 0.6118253746032715,
|
152 |
+
"mean": 0.30591268730163573,
|
153 |
+
"stdev": 0.26405276679992673,
|
154 |
+
"p50": 0.30591268730163573,
|
155 |
+
"p90": 0.5171549007415772,
|
156 |
+
"p95": 0.5435601774215698,
|
157 |
+
"p99": 0.564684398765564,
|
158 |
"values": [
|
159 |
+
0.5699654541015625,
|
160 |
+
0.04185992050170898
|
161 |
]
|
162 |
},
|
163 |
"throughput": {
|
164 |
"unit": "samples/s",
|
165 |
+
"value": 13.07562636673491
|
166 |
},
|
167 |
"energy": null,
|
168 |
"efficiency": null
|
|
|
170 |
"train": {
|
171 |
"memory": {
|
172 |
"unit": "MB",
|
173 |
+
"max_ram": 1116.532736,
|
174 |
+
"max_global_vram": 2945.032192,
|
175 |
+
"max_process_vram": 296709.005312,
|
176 |
"max_reserved": 2497.708032,
|
177 |
"max_allocated": 2195.345408
|
178 |
},
|
179 |
"latency": {
|
180 |
"unit": "s",
|
181 |
"count": 3,
|
182 |
+
"total": 0.12311592102050781,
|
183 |
+
"mean": 0.04103864034016927,
|
184 |
+
"stdev": 0.00010132482928054298,
|
185 |
+
"p50": 0.04101927947998047,
|
186 |
+
"p90": 0.041140879821777346,
|
187 |
+
"p95": 0.041156079864501954,
|
188 |
+
"p99": 0.04116823989868164,
|
189 |
"values": [
|
190 |
+
0.04117127990722656,
|
191 |
+
0.04092536163330078,
|
192 |
+
0.04101927947998047
|
193 |
]
|
194 |
},
|
195 |
"throughput": {
|
196 |
"unit": "samples/s",
|
197 |
+
"value": 146.20367415357828
|
198 |
},
|
199 |
"energy": null,
|
200 |
"efficiency": null
|