Hugging Face's logo

Datasets:
optimum-benchmark
/
rocm

Dataset card Viewer Files Community
IlyasMoutawwakil HF staff commited on
Commit
4fe1e20
·
verified ·
1 Parent(s): 3242e0a

Upload cuda_training_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. cuda_training_transformers_text-generation_openai-community/gpt2/benchmark.json +44 -44
cuda_training_transformers_text-generation_openai-community/gpt2/benchmark.json CHANGED
@@ -86,7 +86,7 @@
86
  "gpu_count": 1,
87
  "gpu_vram_mb": 68702699520,
88
  "optimum_benchmark_version": "0.2.0",
89
- "optimum_benchmark_commit": "b04fb3c909a5873eadf03d7b46ccfac63afcdf9e",
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": 1129.873408,
109
- "max_global_vram": 4308.226048,
110
- "max_process_vram": 381420.56448,
111
  "max_reserved": 2894.06976,
112
  "max_allocated": 2506.73664
113
  },
114
  "latency": {
115
  "unit": "s",
116
  "count": 5,
117
- "total": 0.7523558273315429,
118
- "mean": 0.1504711654663086,
119
- "stdev": 0.21302143192444578,
120
- "p50": 0.0437803840637207,
121
- "p90": 0.3638481796264649,
122
- "p95": 0.47018059005737295,
123
- "p99": 0.5552465184020996,
124
  "values": [
125
- 0.5765130004882812,
126
- 0.04485094833374023,
127
- 0.0437026252746582,
128
- 0.04350886917114258,
129
- 0.0437803840637207
130
  ]
131
  },
132
  "throughput": {
133
  "unit": "samples/s",
134
- "value": 66.45791550168501
135
  },
136
  "energy": null,
137
  "efficiency": null
@@ -139,30 +139,30 @@
139
  "warmup": {
140
  "memory": {
141
  "unit": "MB",
142
- "max_ram": 1129.873408,
143
- "max_global_vram": 4308.226048,
144
- "max_process_vram": 381420.56448,
145
  "max_reserved": 2894.06976,
146
  "max_allocated": 2506.73664
147
  },
148
  "latency": {
149
  "unit": "s",
150
  "count": 2,
151
- "total": 0.6213639488220215,
152
- "mean": 0.31068197441101075,
153
- "stdev": 0.2658310260772705,
154
- "p50": 0.31068197441101075,
155
- "p90": 0.5233467952728271,
156
- "p95": 0.5499298978805541,
157
- "p99": 0.5711963799667358,
158
  "values": [
159
- 0.5765130004882812,
160
- 0.04485094833374023
161
  ]
162
  },
163
  "throughput": {
164
  "unit": "samples/s",
165
- "value": 12.874902084625859
166
  },
167
  "energy": null,
168
  "efficiency": null
@@ -170,31 +170,31 @@
170
  "train": {
171
  "memory": {
172
  "unit": "MB",
173
- "max_ram": 1129.873408,
174
- "max_global_vram": 4308.226048,
175
- "max_process_vram": 381420.56448,
176
  "max_reserved": 2894.06976,
177
  "max_allocated": 2506.73664
178
  },
179
  "latency": {
180
  "unit": "s",
181
  "count": 3,
182
- "total": 0.1309918785095215,
183
- "mean": 0.04366395950317383,
184
- "stdev": 0.0001141676125594552,
185
- "p50": 0.0437026252746582,
186
- "p90": 0.0437648323059082,
187
- "p95": 0.04377260818481445,
188
- "p99": 0.04377882888793945,
189
  "values": [
190
- 0.0437026252746582,
191
- 0.04350886917114258,
192
- 0.0437803840637207
193
  ]
194
  },
195
  "throughput": {
196
  "unit": "samples/s",
197
- "value": 137.41309923035894
198
  },
199
  "energy": null,
200
  "efficiency": null
 
86
  "gpu_count": 1,
87
  "gpu_vram_mb": 68702699520,
88
  "optimum_benchmark_version": "0.2.0",
89
+ "optimum_benchmark_commit": "4ec62071e1c9b9c89fb7e3c044340b391a0c4120",
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": 1129.017344,
109
+ "max_global_vram": 4308.209664,
110
+ "max_process_vram": 356236.087296,
111
  "max_reserved": 2894.06976,
112
  "max_allocated": 2506.73664
113
  },
114
  "latency": {
115
  "unit": "s",
116
  "count": 5,
117
+ "total": 0.7027619667053223,
118
+ "mean": 0.14055239334106445,
119
+ "stdev": 0.19727299135605372,
120
+ "p50": 0.04176374816894531,
121
+ "p90": 0.3381293319702149,
122
+ "p95": 0.4366134330749511,
123
+ "p99": 0.5154007139587402,
124
  "values": [
125
+ 0.5350975341796875,
126
+ 0.04267702865600586,
127
+ 0.04176374816894531,
128
+ 0.04175846862792969,
129
+ 0.04146518707275391
130
  ]
131
  },
132
  "throughput": {
133
  "unit": "samples/s",
134
+ "value": 71.14784574129591
135
  },
136
  "energy": null,
137
  "efficiency": null
 
139
  "warmup": {
140
  "memory": {
141
  "unit": "MB",
142
+ "max_ram": 1129.017344,
143
+ "max_global_vram": 4308.209664,
144
+ "max_process_vram": 356236.087296,
145
  "max_reserved": 2894.06976,
146
  "max_allocated": 2506.73664
147
  },
148
  "latency": {
149
  "unit": "s",
150
  "count": 2,
151
+ "total": 0.5777745628356934,
152
+ "mean": 0.2888872814178467,
153
+ "stdev": 0.24621025276184083,
154
+ "p50": 0.2888872814178467,
155
+ "p90": 0.4858554836273194,
156
+ "p95": 0.5104765089035034,
157
+ "p99": 0.5301733291244507,
158
  "values": [
159
+ 0.5350975341796875,
160
+ 0.04267702865600586
161
  ]
162
  },
163
  "throughput": {
164
  "unit": "samples/s",
165
+ "value": 13.846230891052619
166
  },
167
  "energy": null,
168
  "efficiency": null
 
170
  "train": {
171
  "memory": {
172
  "unit": "MB",
173
+ "max_ram": 1129.017344,
174
+ "max_global_vram": 4308.209664,
175
+ "max_process_vram": 356236.087296,
176
  "max_reserved": 2894.06976,
177
  "max_allocated": 2506.73664
178
  },
179
  "latency": {
180
  "unit": "s",
181
  "count": 3,
182
+ "total": 0.1249874038696289,
183
+ "mean": 0.041662467956542965,
184
+ "stdev": 0.00013951530076131152,
185
+ "p50": 0.04175846862792969,
186
+ "p90": 0.04176269226074219,
187
+ "p95": 0.04176322021484375,
188
+ "p99": 0.041763642578125,
189
  "values": [
190
+ 0.04176374816894531,
191
+ 0.04175846862792969,
192
+ 0.04146518707275391
193
  ]
194
  },
195
  "throughput": {
196
  "unit": "samples/s",
197
+ "value": 144.0145122045685
198
  },
199
  "energy": null,
200
  "efficiency": null