CPU Tasks
LoadModel       945.134 milliseconds
RunComplete     2.19628 seconds
Run     2.08991 seconds
Callbacks       762.3 microseconds, 4 calls, 190.575 microseconds average
Spectrogram     12.2602 milliseconds, 3 calls, 4.08673 milliseconds average
Sample  3.2495 milliseconds, 27 calls, 120.352 microseconds average
Encode  1.31469 seconds
Decode  774.432 milliseconds
DecodeStep      771.17 milliseconds, 27 calls, 28.5618 milliseconds average
    GPU Tasks
LoadModel       803.014 milliseconds
Run     2.07007 seconds
Encode  1.29615 seconds
EncodeLayer     1.11858 seconds, 32 calls, 34.9556 milliseconds average
Decode  773.917 milliseconds
DecodeStep      773.915 milliseconds, 27 calls, 28.6635 milliseconds average
DecodeLayer     719.389 milliseconds, 864 calls, 832.626 microseconds average
    Compute Shaders
mulMatTiled     1.20543 seconds, 705 calls, 1.70983 milliseconds average
mulMatByRowTiled        471.054 milliseconds, 10010 calls, 47.0584 microseconds average
norm    72.3989 milliseconds, 2684 calls, 26.9743 microseconds average
fmaRepeat1      41.7212 milliseconds, 2684 calls, 15.5444 microseconds average
copyTranspose   41.168 milliseconds, 1792 calls, 22.9732 microseconds average
softMaxFixed    40.8861 milliseconds, 896 calls, 45.6318 microseconds average
addRepeatEx     30.1243 milliseconds, 2656 calls, 11.342 microseconds average
copyConvert     28.9217 milliseconds, 1856 calls, 15.5828 microseconds average
softMaxLong     25.3209 milliseconds, 27 calls, 937.811 microseconds average
convolutionMain2Fixed   19.8769 milliseconds
addRepeatScale  18.2236 milliseconds, 1728 calls, 10.5461 microseconds average
addRepeatGelu   15.7554 milliseconds, 898 calls, 17.545 microseconds average
addRepeat       14.2968 milliseconds, 960 calls, 14.8925 microseconds average
scaleInPlace    13.9332 milliseconds, 896 calls, 15.5504 microseconds average
softMax 8.5928 milliseconds, 864 calls, 9.94537 microseconds average
convolutionMain 8.532 milliseconds
diagMaskInf     5.6745 milliseconds, 864 calls, 6.56771 microseconds average
convolutionPrep1        2.303 milliseconds, 2 calls, 1.1515 milliseconds average
convolutionPrep2        422.9 microseconds, 2 calls, 211.45 microseconds average
addRows 198.7 microseconds, 27 calls, 7.35926 microseconds average
add     119.8 microseconds
    Memory Usage
Model   892.591 KB RAM, 2.8815 GB VRAM
Context 1.98376 MB RAM, 1.07641 GB VRAM
Total   2.85543 MB RAM, 3.95791 GB VRAM