CPU Tasks
LoadModel       1.99675 seconds
RunComplete     81.256 seconds
Run     81.1666 seconds
Callbacks       17.8976 milliseconds, 37 calls, 483.719 microseconds average
Spectrogram     483.273 milliseconds, 42 calls, 11.5065 milliseconds average
Sample  140.511 milliseconds, 511 calls, 274.972 microseconds average
Encode  50.3768 seconds, 10 calls, 5.03768 seconds average
Decode  30.7646 seconds, 10 calls, 3.07646 seconds average
DecodeStep      30.6234 seconds, 511 calls, 59.9284 milliseconds average
    GPU Tasks
LoadModel       976.318 milliseconds
Run     80.8284 seconds
Encode  51.1656 seconds, 10 calls, 5.11656 seconds average
EncodeLayer     43.8924 seconds, 240 calls, 182.885 milliseconds average
Decode  29.6502 seconds, 10 calls, 2.96502 seconds average
DecodeStep      29.6441 seconds, 511 calls, 58.012 milliseconds average
DecodeLayer     26.9439 seconds, 12264 calls, 2.19699 milliseconds average
    Compute Shaders
mulMatTiledEx   37.1919 seconds, 2400 calls, 15.4966 milliseconds average
mulMatTiled     13.9953 seconds, 2890 calls, 4.84268 milliseconds average
mulMatByRowTiled        11.8792 seconds, 120741 calls, 98.3858 microseconds average
mulMatByRowTiledEx      4.47094 seconds, 24048 calls, 185.917 microseconds average
softMaxFixed    2.44162 seconds, 12504 calls, 195.267 microseconds average
convolutionMain2Fixed   1.51096 seconds, 10 calls, 151.096 milliseconds average
matReshapePanels        1.38964 seconds, 1450 calls, 958.371 microseconds average
addRepeatGelu   963.292 milliseconds, 12524 calls, 76.9157 microseconds average
normFixed       925.912 milliseconds, 37793 calls, 24.4996 microseconds average
copyConvert     875.162 milliseconds, 25488 calls, 34.3362 microseconds average
scaleInPlace    770.121 milliseconds, 12504 calls, 61.59 microseconds average
fmaRepeat1      696.227 milliseconds, 37793 calls, 18.4221 microseconds average
copyTranspose   657.921 milliseconds, 25008 calls, 26.3084 microseconds average
addRepeatEx     630.019 milliseconds, 37272 calls, 16.9033 microseconds average
softMaxLong     623.51 milliseconds, 511 calls, 1.22018 milliseconds average
convolutionMain 471.348 milliseconds, 10 calls, 47.1348 milliseconds average
addRepeatScale  379.836 milliseconds, 24528 calls, 15.4858 microseconds average
addRepeat       354.984 milliseconds, 12984 calls, 27.3401 microseconds average
softMax 197.387 milliseconds, 12264 calls, 16.0948 microseconds average
diagMaskInf     131.012 milliseconds, 12264 calls, 10.6827 microseconds average
convolutionPrep2        49.7619 milliseconds, 20 calls, 2.48809 milliseconds average
convolutionPrep1        42.2907 milliseconds, 20 calls, 2.11454 milliseconds average
add     10.5473 milliseconds, 10 calls, 1.05473 milliseconds average
addRows 2.1075 milliseconds, 511 calls, 4.12427 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