Hugging Face's logo

Datasets:
optimum-benchmark
/
rocm

Dataset card Viewer Files Community
IlyasMoutawwakil HF staff commited on
Commit
bb04528
·
verified ·
1 Parent(s): 2439d90

Upload cuda_training_transformers_text-classification_FacebookAI/roberta-base/benchmark.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. cuda_training_transformers_text-classification_FacebookAI/roberta-base/benchmark.json +44 -44
cuda_training_transformers_text-classification_FacebookAI/roberta-base/benchmark.json CHANGED
@@ -86,7 +86,7 @@
86
  "gpu_count": 1,
87
  "gpu_vram_mb": 68702699520,
88
  "optimum_benchmark_version": "0.2.1",
89
- "optimum_benchmark_commit": "c1d0b062e90b79e7705510c58cea731c0d90da8a",
90
  "transformers_version": "4.40.2",
91
  "transformers_commit": null,
92
  "accelerate_version": "0.30.1",
@@ -105,33 +105,33 @@
105
  "overall": {
106
  "memory": {
107
  "unit": "MB",
108
- "max_ram": 1124.839424,
109
- "max_global_vram": 4135.804928,
110
- "max_process_vram": 290264.989696,
111
  "max_reserved": 2707.423232,
112
  "max_allocated": 2497.900032
113
  },
114
  "latency": {
115
  "unit": "s",
116
  "count": 5,
117
- "total": 0.6849701194763185,
118
- "mean": 0.1369940238952637,
119
- "stdev": 0.18969544167240582,
120
- "p50": 0.04213109970092774,
121
- "p90": 0.3269433593750001,
122
- "p95": 0.42166381835937494,
123
- "p99": 0.497440185546875,
124
  "values": [
125
- 0.51638427734375,
126
- 0.042781982421875,
127
- 0.04181381988525391,
128
- 0.04213109970092774,
129
- 0.04185894012451172
130
  ]
131
  },
132
  "throughput": {
133
  "unit": "samples/s",
134
- "value": 72.99588489820052
135
  },
136
  "energy": null,
137
  "efficiency": null
@@ -139,30 +139,30 @@
139
  "warmup": {
140
  "memory": {
141
  "unit": "MB",
142
- "max_ram": 1124.839424,
143
- "max_global_vram": 4135.804928,
144
- "max_process_vram": 290264.989696,
145
  "max_reserved": 2707.423232,
146
  "max_allocated": 2497.900032
147
  },
148
  "latency": {
149
  "unit": "s",
150
  "count": 2,
151
- "total": 0.5591662597656251,
152
- "mean": 0.27958312988281253,
153
- "stdev": 0.23680114746093753,
154
- "p50": 0.27958312988281253,
155
- "p90": 0.46902404785156254,
156
- "p95": 0.49270416259765626,
157
- "p99": 0.5116482543945313,
158
  "values": [
159
- 0.51638427734375,
160
- 0.042781982421875
161
  ]
162
  },
163
  "throughput": {
164
  "unit": "samples/s",
165
- "value": 14.307014882037421
166
  },
167
  "energy": null,
168
  "efficiency": null
@@ -170,31 +170,31 @@
170
  "train": {
171
  "memory": {
172
  "unit": "MB",
173
- "max_ram": 1124.839424,
174
- "max_global_vram": 4135.804928,
175
- "max_process_vram": 290264.989696,
176
  "max_reserved": 2707.423232,
177
  "max_allocated": 2497.900032
178
  },
179
  "latency": {
180
  "unit": "s",
181
  "count": 3,
182
- "total": 0.12580385971069336,
183
- "mean": 0.04193461990356445,
184
- "stdev": 0.00014014799814626502,
185
- "p50": 0.04185894012451172,
186
- "p90": 0.042076667785644534,
187
- "p95": 0.042103883743286136,
188
- "p99": 0.042125656509399416,
189
  "values": [
190
- 0.04181381988525391,
191
- 0.04213109970092774,
192
- 0.04185894012451172
193
  ]
194
  },
195
  "throughput": {
196
  "unit": "samples/s",
197
- "value": 143.0798708512915
198
  },
199
  "energy": null,
200
  "efficiency": null
 
86
  "gpu_count": 1,
87
  "gpu_vram_mb": 68702699520,
88
  "optimum_benchmark_version": "0.2.1",
89
+ "optimum_benchmark_commit": "48414f58841d7ba7c7fd42d74fd524d1d23c3081",
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": 1121.91488,
109
+ "max_global_vram": 4135.821312,
110
+ "max_process_vram": 305536.049152,
111
  "max_reserved": 2707.423232,
112
  "max_allocated": 2497.900032
113
  },
114
  "latency": {
115
  "unit": "s",
116
  "count": 5,
117
+ "total": 0.777501682281494,
118
+ "mean": 0.1555003364562988,
119
+ "stdev": 0.2267338792443409,
120
+ "p50": 0.04209430694580078,
121
+ "p90": 0.38246746673583987,
122
+ "p95": 0.4957175590515135,
123
+ "p99": 0.5863176329040527,
124
  "values": [
125
+ 0.6089676513671874,
126
+ 0.04271718978881836,
127
+ 0.04209430694580078,
128
+ 0.04178246688842773,
129
+ 0.041940067291259764
130
  ]
131
  },
132
  "throughput": {
133
  "unit": "samples/s",
134
+ "value": 64.30854252723988
135
  },
136
  "energy": null,
137
  "efficiency": null
 
139
  "warmup": {
140
  "memory": {
141
  "unit": "MB",
142
+ "max_ram": 1121.91488,
143
+ "max_global_vram": 4135.821312,
144
+ "max_process_vram": 305536.049152,
145
  "max_reserved": 2707.423232,
146
  "max_allocated": 2497.900032
147
  },
148
  "latency": {
149
  "unit": "s",
150
  "count": 2,
151
+ "total": 0.6516848411560058,
152
+ "mean": 0.3258424205780029,
153
+ "stdev": 0.28312523078918456,
154
+ "p50": 0.3258424205780029,
155
+ "p90": 0.5523426052093505,
156
+ "p95": 0.5806551282882689,
157
+ "p99": 0.6033051467514038,
158
  "values": [
159
+ 0.6089676513671874,
160
+ 0.04271718978881836
161
  ]
162
  },
163
  "throughput": {
164
  "unit": "samples/s",
165
+ "value": 12.275872468982124
166
  },
167
  "energy": null,
168
  "efficiency": null
 
170
  "train": {
171
  "memory": {
172
  "unit": "MB",
173
+ "max_ram": 1121.91488,
174
+ "max_global_vram": 4135.821312,
175
+ "max_process_vram": 305536.049152,
176
  "max_reserved": 2707.423232,
177
  "max_allocated": 2497.900032
178
  },
179
  "latency": {
180
  "unit": "s",
181
  "count": 3,
182
+ "total": 0.12581684112548827,
183
+ "mean": 0.041938947041829426,
184
+ "stdev": 0.00012731063470817258,
185
+ "p50": 0.041940067291259764,
186
+ "p90": 0.04206345901489258,
187
+ "p95": 0.042078882980346684,
188
+ "p99": 0.04209122215270996,
189
  "values": [
190
+ 0.04209430694580078,
191
+ 0.04178246688842773,
192
+ 0.041940067291259764
193
  ]
194
  },
195
  "throughput": {
196
  "unit": "samples/s",
197
+ "value": 143.06510828742717
198
  },
199
  "energy": null,
200
  "efficiency": null