]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
whisper : fix "bench-all outputs an invalid result on larger models" (#3002)
authorFujimoto Seiji <redacted>
Fri, 4 Apr 2025 15:36:19 +0000 (00:36 +0900)
committerGitHub <redacted>
Fri, 4 Apr 2025 15:36:19 +0000 (18:36 +0300)
commite6234cd435eb1299ac405916f4083a62ac044b81
tree486e3193efe207a279d883e54e02f78afbafda44
parent2b6d0d22001b726fc7ecce2357010ff4d37af19c
whisper : fix "bench-all outputs an invalid result on larger models" (#3002)

The benchmark script 'scripts/bench-all.sh' assumes that the 11th
field of the output line is a timestamp. This assumption does not
hold when the target model takes a bit longer to process.

Fix this issue by introducing an explicit whitespace to the output
lines of `whisper_print_timings()`.

Signed-off-by: Fujimoto Seiji <redacted>
src/whisper.cpp