Upload cuda_training_transformers_image-classification_google/vit-base-patch16-224/benchmark.json with huggingface_hub
Browse files
cuda_training_transformers_image-classification_google/vit-base-patch16-224/benchmark.json
CHANGED
@@ -89,7 +89,7 @@
|
|
89 |
"gpu_count": 1,
|
90 |
"gpu_vram_mb": 68702699520,
|
91 |
"optimum_benchmark_version": "0.2.1",
|
92 |
-
"optimum_benchmark_commit": "
|
93 |
"transformers_version": "4.42.3",
|
94 |
"transformers_commit": null,
|
95 |
"accelerate_version": "0.31.0",
|
@@ -108,33 +108,33 @@
|
|
108 |
"overall": {
|
109 |
"memory": {
|
110 |
"unit": "MB",
|
111 |
-
"max_ram":
|
112 |
-
"max_global_vram": 2427.
|
113 |
-
"max_process_vram":
|
114 |
"max_reserved": 1935.671296,
|
115 |
"max_allocated": 1738.272256
|
116 |
},
|
117 |
"latency": {
|
118 |
"unit": "s",
|
119 |
"count": 5,
|
120 |
-
"total":
|
121 |
-
"mean": 0.
|
122 |
-
"stdev": 1.
|
123 |
-
"p50": 0.
|
124 |
-
"p90": 1.
|
125 |
-
"p95": 2.
|
126 |
-
"p99":
|
127 |
"values": [
|
128 |
-
|
129 |
-
0.
|
130 |
-
0.
|
131 |
-
0.
|
132 |
-
0.
|
133 |
]
|
134 |
},
|
135 |
"throughput": {
|
136 |
"unit": "samples/s",
|
137 |
-
"value":
|
138 |
},
|
139 |
"energy": null,
|
140 |
"efficiency": null
|
@@ -142,30 +142,30 @@
|
|
142 |
"warmup": {
|
143 |
"memory": {
|
144 |
"unit": "MB",
|
145 |
-
"max_ram":
|
146 |
-
"max_global_vram": 2427.
|
147 |
-
"max_process_vram":
|
148 |
"max_reserved": 1935.671296,
|
149 |
"max_allocated": 1738.272256
|
150 |
},
|
151 |
"latency": {
|
152 |
"unit": "s",
|
153 |
"count": 2,
|
154 |
-
"total":
|
155 |
-
"mean": 1.
|
156 |
-
"stdev": 1.
|
157 |
-
"p50": 1.
|
158 |
-
"p90": 2.
|
159 |
-
"p95": 2.
|
160 |
-
"p99":
|
161 |
"values": [
|
162 |
-
|
163 |
-
0.
|
164 |
]
|
165 |
},
|
166 |
"throughput": {
|
167 |
"unit": "samples/s",
|
168 |
-
"value": 2.
|
169 |
},
|
170 |
"energy": null,
|
171 |
"efficiency": null
|
@@ -173,31 +173,31 @@
|
|
173 |
"train": {
|
174 |
"memory": {
|
175 |
"unit": "MB",
|
176 |
-
"max_ram":
|
177 |
-
"max_global_vram": 2427.
|
178 |
-
"max_process_vram":
|
179 |
"max_reserved": 1935.671296,
|
180 |
"max_allocated": 1738.272256
|
181 |
},
|
182 |
"latency": {
|
183 |
"unit": "s",
|
184 |
"count": 3,
|
185 |
-
"total": 0.
|
186 |
-
"mean": 0.
|
187 |
-
"stdev": 0.
|
188 |
-
"p50": 0.
|
189 |
-
"p90": 0.
|
190 |
-
"p95": 0.
|
191 |
-
"p99": 0.
|
192 |
"values": [
|
193 |
-
0.
|
194 |
-
0.
|
195 |
-
0.
|
196 |
]
|
197 |
},
|
198 |
"throughput": {
|
199 |
"unit": "samples/s",
|
200 |
-
"value":
|
201 |
},
|
202 |
"energy": null,
|
203 |
"efficiency": null
|
|
|
89 |
"gpu_count": 1,
|
90 |
"gpu_vram_mb": 68702699520,
|
91 |
"optimum_benchmark_version": "0.2.1",
|
92 |
+
"optimum_benchmark_commit": "0d561969d04a7091baf8691bc894eb753a189d4f",
|
93 |
"transformers_version": "4.42.3",
|
94 |
"transformers_commit": null,
|
95 |
"accelerate_version": "0.31.0",
|
|
|
108 |
"overall": {
|
109 |
"memory": {
|
110 |
"unit": "MB",
|
111 |
+
"max_ram": 2577.682432,
|
112 |
+
"max_global_vram": 2427.838464,
|
113 |
+
"max_process_vram": 662334.09536,
|
114 |
"max_reserved": 1935.671296,
|
115 |
"max_allocated": 1738.272256
|
116 |
},
|
117 |
"latency": {
|
118 |
"unit": "s",
|
119 |
"count": 5,
|
120 |
+
"total": 3.2920257453918453,
|
121 |
+
"mean": 0.6584051490783691,
|
122 |
+
"stdev": 1.2364248515128586,
|
123 |
+
"p50": 0.04115587615966797,
|
124 |
+
"p90": 1.8953758468627933,
|
125 |
+
"p95": 2.5133148765563957,
|
126 |
+
"p99": 3.0076661003112792,
|
127 |
"values": [
|
128 |
+
3.13125390625,
|
129 |
+
0.039442920684814455,
|
130 |
+
0.041558757781982425,
|
131 |
+
0.04115587615966797,
|
132 |
+
0.03861428451538086
|
133 |
]
|
134 |
},
|
135 |
"throughput": {
|
136 |
"unit": "samples/s",
|
137 |
+
"value": 15.188216577585898
|
138 |
},
|
139 |
"energy": null,
|
140 |
"efficiency": null
|
|
|
142 |
"warmup": {
|
143 |
"memory": {
|
144 |
"unit": "MB",
|
145 |
+
"max_ram": 2577.682432,
|
146 |
+
"max_global_vram": 2427.838464,
|
147 |
+
"max_process_vram": 662334.09536,
|
148 |
"max_reserved": 1935.671296,
|
149 |
"max_allocated": 1738.272256
|
150 |
},
|
151 |
"latency": {
|
152 |
"unit": "s",
|
153 |
"count": 2,
|
154 |
+
"total": 3.1706968269348144,
|
155 |
+
"mean": 1.5853484134674072,
|
156 |
+
"stdev": 1.5459054927825928,
|
157 |
+
"p50": 1.5853484134674072,
|
158 |
+
"p90": 2.8220728076934813,
|
159 |
+
"p95": 2.9766633569717404,
|
160 |
+
"p99": 3.100335796394348,
|
161 |
"values": [
|
162 |
+
3.13125390625,
|
163 |
+
0.039442920684814455
|
164 |
]
|
165 |
},
|
166 |
"throughput": {
|
167 |
"unit": "samples/s",
|
168 |
+
"value": 2.5231046790852547
|
169 |
},
|
170 |
"energy": null,
|
171 |
"efficiency": null
|
|
|
173 |
"train": {
|
174 |
"memory": {
|
175 |
"unit": "MB",
|
176 |
+
"max_ram": 2577.682432,
|
177 |
+
"max_global_vram": 2427.838464,
|
178 |
+
"max_process_vram": 662334.09536,
|
179 |
"max_reserved": 1935.671296,
|
180 |
"max_allocated": 1738.272256
|
181 |
},
|
182 |
"latency": {
|
183 |
"unit": "s",
|
184 |
"count": 3,
|
185 |
+
"total": 0.12132891845703125,
|
186 |
+
"mean": 0.040442972819010416,
|
187 |
+
"stdev": 0.001303496345544894,
|
188 |
+
"p50": 0.04115587615966797,
|
189 |
+
"p90": 0.04147818145751953,
|
190 |
+
"p95": 0.04151846961975098,
|
191 |
+
"p99": 0.041550700149536136,
|
192 |
"values": [
|
193 |
+
0.041558757781982425,
|
194 |
+
0.04115587615966797,
|
195 |
+
0.03861428451538086
|
196 |
]
|
197 |
},
|
198 |
"throughput": {
|
199 |
"unit": "samples/s",
|
200 |
+
"value": 148.3570465220517
|
201 |
},
|
202 |
"energy": null,
|
203 |
"efficiency": null
|