IlyasMoutawwakil HF Staff commited on
Commit
bb176f5
·
verified ·
1 Parent(s): 0fa6eb8

Upload cuda_training_transformers_fill-mask_hf-internal-testing/tiny-random-BertModel/benchmark.json with huggingface_hub

Browse files
cuda_training_transformers_fill-mask_hf-internal-testing/tiny-random-BertModel/benchmark.json CHANGED
@@ -3,7 +3,7 @@
3
  "name": "cuda_training_transformers_fill-mask_hf-internal-testing/tiny-random-BertModel",
4
  "backend": {
5
  "name": "pytorch",
6
- "version": "2.7.0+rocm6.3",
7
  "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
8
  "model": "hf-internal-testing/tiny-random-BertModel",
9
  "processor": "hf-internal-testing/tiny-random-BertModel",
@@ -11,7 +11,7 @@
11
  "library": "transformers",
12
  "model_type": "bert",
13
  "device": "cuda",
14
- "device_ids": "5",
15
  "seed": 42,
16
  "inter_op_num_threads": null,
17
  "intra_op_num_threads": null,
@@ -109,7 +109,7 @@
109
  "optimum_commit": null,
110
  "timm_version": "1.0.15",
111
  "timm_commit": null,
112
- "peft_version": null,
113
  "peft_commit": null
114
  },
115
  "print_report": true,
@@ -119,34 +119,34 @@
119
  "overall": {
120
  "memory": {
121
  "unit": "MB",
122
- "max_ram": 2317.877248,
123
  "max_global_vram": 68702.69952,
124
- "max_process_vram": 0.0,
125
- "max_reserved": 81.788928,
126
- "max_allocated": 81.408512
127
  },
128
  "latency": {
129
  "unit": "s",
130
  "values": [
131
- 0.9848811645507812,
132
- 0.016590362548828126,
133
- 0.01589116382598877,
134
- 0.015356125831604004,
135
- 0.015837565422058105
136
  ],
137
  "count": 5,
138
- "total": 1.0485563821792603,
139
- "mean": 0.20971127643585205,
140
- "p50": 0.01589116382598877,
141
- "p90": 0.5975648437500001,
142
- "p95": 0.7912230041503905,
143
- "p99": 0.9461495324707031,
144
- "stdev": 0.3875851440035934,
145
- "stdev_": 184.818456399101
146
  },
147
  "throughput": {
148
  "unit": "samples/s",
149
- "value": 47.684607952204566
150
  },
151
  "energy": null,
152
  "efficiency": null
@@ -154,31 +154,31 @@
154
  "warmup": {
155
  "memory": {
156
  "unit": "MB",
157
- "max_ram": 2317.877248,
158
  "max_global_vram": 68702.69952,
159
- "max_process_vram": 0.0,
160
- "max_reserved": 81.788928,
161
- "max_allocated": 81.408512
162
  },
163
  "latency": {
164
  "unit": "s",
165
  "values": [
166
- 0.9848811645507812,
167
- 0.016590362548828126
168
  ],
169
  "count": 2,
170
- "total": 1.0014715270996093,
171
- "mean": 0.5007357635498046,
172
- "p50": 0.5007357635498046,
173
- "p90": 0.8880520843505859,
174
- "p95": 0.9364666244506835,
175
- "p99": 0.9751982565307616,
176
- "stdev": 0.4841454010009765,
177
- "stdev_": 96.68680294948057
178
  },
179
  "throughput": {
180
  "unit": "samples/s",
181
- "value": 7.988245080885157
182
  },
183
  "energy": null,
184
  "efficiency": null
@@ -186,32 +186,32 @@
186
  "train": {
187
  "memory": {
188
  "unit": "MB",
189
- "max_ram": 2317.877248,
190
  "max_global_vram": 68702.69952,
191
- "max_process_vram": 0.0,
192
- "max_reserved": 81.788928,
193
- "max_allocated": 81.408512
194
  },
195
  "latency": {
196
  "unit": "s",
197
  "values": [
198
- 0.01589116382598877,
199
- 0.015356125831604004,
200
- 0.015837565422058105
201
  ],
202
  "count": 3,
203
- "total": 0.04708485507965088,
204
- "mean": 0.015694951693216958,
205
- "p50": 0.015837565422058105,
206
- "p90": 0.01588044414520264,
207
- "p95": 0.015885803985595703,
208
- "p99": 0.015890091857910158,
209
- "stdev": 0.00024058320888942626,
210
- "stdev_": 1.5328700182836592
211
  },
212
  "throughput": {
213
  "unit": "samples/s",
214
- "value": 382.2885292850617
215
  },
216
  "energy": null,
217
  "efficiency": null
 
3
  "name": "cuda_training_transformers_fill-mask_hf-internal-testing/tiny-random-BertModel",
4
  "backend": {
5
  "name": "pytorch",
6
+ "version": "2.3.1+rocm5.7",
7
  "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
8
  "model": "hf-internal-testing/tiny-random-BertModel",
9
  "processor": "hf-internal-testing/tiny-random-BertModel",
 
11
  "library": "transformers",
12
  "model_type": "bert",
13
  "device": "cuda",
14
+ "device_ids": "6",
15
  "seed": 42,
16
  "inter_op_num_threads": null,
17
  "intra_op_num_threads": null,
 
109
  "optimum_commit": null,
110
  "timm_version": "1.0.15",
111
  "timm_commit": null,
112
+ "peft_version": "0.15.2",
113
  "peft_commit": null
114
  },
115
  "print_report": true,
 
119
  "overall": {
120
  "memory": {
121
  "unit": "MB",
122
+ "max_ram": 1282.916352,
123
  "max_global_vram": 68702.69952,
124
+ "max_process_vram": 59706.052608,
125
+ "max_reserved": 2.097152,
126
+ "max_allocated": 1.940992
127
  },
128
  "latency": {
129
  "unit": "s",
130
  "values": [
131
+ 0.6948958740234376,
132
+ 0.015948529243469237,
133
+ 0.014694604873657226,
134
+ 0.014702765464782715,
135
+ 0.014656845092773438
136
  ],
137
  "count": 5,
138
+ "total": 0.7548986186981201,
139
+ "mean": 0.15097972373962404,
140
+ "p50": 0.014702765464782715,
141
+ "p90": 0.42331693611145027,
142
+ "p95": 0.5591064050674438,
143
+ "p99": 0.6677379802322388,
144
+ "stdev": 0.27195851604615723,
145
+ "stdev_": 180.12916523490946
146
  },
147
  "throughput": {
148
  "unit": "samples/s",
149
+ "value": 66.2340594638109
150
  },
151
  "energy": null,
152
  "efficiency": null
 
154
  "warmup": {
155
  "memory": {
156
  "unit": "MB",
157
+ "max_ram": 1282.916352,
158
  "max_global_vram": 68702.69952,
159
+ "max_process_vram": 59706.052608,
160
+ "max_reserved": 2.097152,
161
+ "max_allocated": 1.940992
162
  },
163
  "latency": {
164
  "unit": "s",
165
  "values": [
166
+ 0.6948958740234376,
167
+ 0.015948529243469237
168
  ],
169
  "count": 2,
170
+ "total": 0.7108444032669068,
171
+ "mean": 0.3554222016334534,
172
+ "p50": 0.3554222016334534,
173
+ "p90": 0.6270011395454407,
174
+ "p95": 0.6609485067844391,
175
+ "p99": 0.6881064005756379,
176
+ "stdev": 0.33947367238998416,
177
+ "stdev_": 95.51279318788393
178
  },
179
  "throughput": {
180
  "unit": "samples/s",
181
+ "value": 11.254220984555142
182
  },
183
  "energy": null,
184
  "efficiency": null
 
186
  "train": {
187
  "memory": {
188
  "unit": "MB",
189
+ "max_ram": 1282.916352,
190
  "max_global_vram": 68702.69952,
191
+ "max_process_vram": 59706.052608,
192
+ "max_reserved": 2.097152,
193
+ "max_allocated": 1.940992
194
  },
195
  "latency": {
196
  "unit": "s",
197
  "values": [
198
+ 0.014694604873657226,
199
+ 0.014702765464782715,
200
+ 0.014656845092773438
201
  ],
202
  "count": 3,
203
+ "total": 0.04405421543121338,
204
+ "mean": 0.014684738477071127,
205
+ "p50": 0.014694604873657226,
206
+ "p90": 0.014701133346557618,
207
+ "p95": 0.014701949405670166,
208
+ "p99": 0.014702602252960205,
209
+ "stdev": 2.000299106905853e-05,
210
+ "stdev_": 0.1362161886661541
211
  },
212
  "throughput": {
213
  "unit": "samples/s",
214
+ "value": 408.58746033294705
215
  },
216
  "energy": null,
217
  "efficiency": null