CPU Tasks
LoadModel       841.605 milliseconds
RunComplete     62.1145 seconds
Run     62.0268 seconds
Callbacks       10.184 milliseconds, 37 calls, 275.243 microseconds average
Spectrogram     200.241 milliseconds, 42 calls, 4.76764 milliseconds average
Sample  63.0473 milliseconds, 511 calls, 123.38 microseconds average
Encode  37.2409 seconds, 10 calls, 3.72409 seconds average
Decode  24.7715 seconds, 10 calls, 2.47715 seconds average
DecodeStep      24.7082 seconds, 511 calls, 48.3526 milliseconds average
    GPU Tasks
LoadModel       410.579 milliseconds
Run     61.8044 seconds
Encode  37.8702 seconds, 10 calls, 3.78702 seconds average
EncodeLayer     32.1896 seconds, 240 calls, 134.123 milliseconds average
Decode  23.9262 seconds, 10 calls, 2.39262 seconds average
DecodeStep      23.9233 seconds, 511 calls, 46.8167 milliseconds average
DecodeLayer     21.6888 seconds, 12264 calls, 1.76849 milliseconds average
    Compute Shaders
mulMatTiledEx   27.5216 seconds, 2400 calls, 11.4673 milliseconds average
mulMatTiled     10.2385 seconds, 2890 calls, 3.54273 milliseconds average
mulMatByRowTiled        9.38114 seconds, 120741 calls, 77.6964 microseconds average
mulMatByRowTiledEx      4.19991 seconds, 24048 calls, 174.647 microseconds average
softMaxFixed    1.95105 seconds, 12504 calls, 156.034 microseconds average
convolutionMain2Fixed   1.31354 seconds, 10 calls, 131.354 milliseconds average
matReshapePanels        1.04699 seconds, 1450 calls, 722.064 microseconds average
addRepeatGelu   777.683 milliseconds, 12524 calls, 62.0954 microseconds average
scaleInPlace    750.056 milliseconds, 12504 calls, 59.9853 microseconds average
copyConvert     701.517 milliseconds, 25488 calls, 27.5234 microseconds average
normFixed       697.931 milliseconds, 37793 calls, 18.4672 microseconds average
fmaRepeat1      529.007 milliseconds, 37793 calls, 13.9975 microseconds average
addRepeatEx     511.269 milliseconds, 37272 calls, 13.7172 microseconds average
copyTranspose   459.017 milliseconds, 25008 calls, 18.3548 microseconds average
softMaxLong     382.205 milliseconds, 511 calls, 747.955 microseconds average
convolutionMain 328.996 milliseconds, 10 calls, 32.8996 milliseconds average
addRepeat       305.31 milliseconds, 12984 calls, 23.5144 microseconds average
addRepeatScale  261.749 milliseconds, 24528 calls, 10.6715 microseconds average
softMax 148.894 milliseconds, 12264 calls, 12.1407 microseconds average
diagMaskInf     104.681 milliseconds, 12264 calls, 8.53562 microseconds average
convolutionPrep2        45.8033 milliseconds, 20 calls, 2.29017 milliseconds average
convolutionPrep1        32.3779 milliseconds, 20 calls, 1.61889 milliseconds average
add     7.3228 milliseconds, 10 calls, 732.28 microseconds average
addRows 1.6948 milliseconds, 511 calls, 3.31663 microseconds average
    Memory Usage
Model   877.966 KB RAM, 1.42785 GB VRAM
Context 91.0716 MB RAM, 833.407 MB VRAM
Total   91.929 MB RAM, 2.24172 GB VRAM