vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

add tokens per second output (#246)

Browse source 
* add tokens per second output

* Update gpttype_adapter.cpp

simplify

---------

Co-authored-by: LostRuins <39025047+LostRuins@users.noreply.github.com>
This commit is contained in:
YellowRoseCx 2023-06-17 06:54:29 -05:00 • committed by GitHub
parent b1b8dc32c9
commit 971fe9f007
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

3
gpttype_adapter.cpp
View file

@ -1280,7 +1280,8 @@ generation_outputs gpttype_generate(const generation_inputs inputs, generation_o
float pt1 = (time1*1000.0/(embd_inp.size()==0?1:embd_inp.size()));
int realnpredict = params.n_predict-stopper_unused_tokens;
float pt2 = (time2*1000.0/(realnpredict==0?1:realnpredict));
printf("\nTime Taken - Processing:%.1fs (%.0fms/T), Generation:%.1fs (%.0fms/T), Total:%.1fs", time1, pt1, time2, pt2, (time1 + time2));
float tokens_per_second = (realnpredict == 0 ? 0 : realnpredict / (time1 + time2));
printf("\nTime Taken - Processing:%.1fs (%.0fms/T), Generation:%.1fs (%.0fms/T), Total:%.1fs (%.1fT/s)", time1, pt1, time2, pt2, (time1 + time2), tokens_per_second);
fflush(stdout);
output.status = 1;
generation_finished = true;