Commit 60fa4b1
authored
Update benchmark of gpt2 model with past state (microsoft#4043)
* update benchmark_gpt2 to use past state only
* update dynamic axes of input/output tensors
* Remove --use_openmp option since it is default for onnxruntime 1.3 cpu.
* Use same option names as benchmark.py1 parent ed0a8e5 commit 60fa4b1
2 files changed
Lines changed: 163 additions & 117 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
224 | 224 | | |
225 | 225 | | |
226 | 226 | | |
227 | | - | |
| 227 | + | |
228 | 228 | | |
229 | 229 | | |
230 | 230 | | |
| |||
263 | 263 | | |
264 | 264 | | |
265 | 265 | | |
266 | | - | |
| 266 | + | |
| 267 | + | |
267 | 268 | | |
268 | 269 | | |
269 | 270 | | |
| |||
0 commit comments