CPU Tasks LoadModel 939.886 milliseconds RunComplete 48.7479 seconds Run 48.6305 seconds Callbacks 10.5582 milliseconds, 37 calls, 285.357 microseconds average Spectrogram 280.966 milliseconds, 42 calls, 6.68965 milliseconds average Sample 65.5797 milliseconds, 511 calls, 128.336 microseconds average Encode 19.0653 seconds, 10 calls, 1.90653 seconds average Decode 29.5369 seconds, 10 calls, 2.95369 seconds average DecodeStep 29.4709 seconds, 511 calls, 57.6731 milliseconds average GPU Tasks LoadModel 586.498 milliseconds Run 48.4589 seconds Encode 19.2258 seconds, 10 calls, 1.92258 seconds average EncodeLayer 15.7109 seconds, 240 calls, 65.4622 milliseconds average Decode 29.233 seconds, 10 calls, 2.9233 seconds average DecodeStep 29.233 seconds, 511 calls, 57.2074 milliseconds average DecodeLayer 27.6558 seconds, 12264 calls, 2.25504 milliseconds average Compute Shaders mulMatTiled 19.1596 seconds, 5290 calls, 3.62186 milliseconds average mulMatByRowTiled 12.681 seconds, 144789 calls, 87.5829 microseconds average fmaRepeat1 3.10945 seconds, 37793 calls, 82.2758 microseconds average copyTranspose 2.83737 seconds, 25008 calls, 113.458 microseconds average softMaxFixed 1.6294 seconds, 12504 calls, 130.31 microseconds average addRepeatScale 1.54396 seconds, 24528 calls, 62.9467 microseconds average addRepeatEx 1.06992 seconds, 37272 calls, 28.7056 microseconds average normFixed 1.06753 seconds, 37793 calls, 28.2467 microseconds average copyConvert 994.495 milliseconds, 25488 calls, 39.0182 microseconds average convolutionMain2Fixed 954.715 milliseconds, 10 calls, 95.4715 milliseconds average softMax 742.126 milliseconds, 12264 calls, 60.5126 microseconds average addRepeatGelu 506.056 milliseconds, 12524 calls, 40.4069 microseconds average softMaxLong 491.226 milliseconds, 511 calls, 961.303 microseconds average scaleInPlace 438.956 milliseconds, 12504 calls, 35.1053 microseconds average addRepeat 403.997 milliseconds, 12984 calls, 31.1149 microseconds average diagMaskInf 366.713 milliseconds, 12264 calls, 29.9016 microseconds average convolutionMain 276.364 milliseconds, 10 calls, 27.6364 milliseconds average convolutionPrep1 44.9126 milliseconds, 20 calls, 2.24563 milliseconds average convolutionPrep2 20.0013 milliseconds, 20 calls, 1.00006 milliseconds average addRows 7.2369 milliseconds, 511 calls, 14.1622 microseconds average add 2.453 milliseconds, 10 calls, 245.3 microseconds average Memory Usage Model 877.966 KB RAM, 1.42785 GB VRAM Context 91.0716 MB RAM, 785.219 MB VRAM Total 91.929 MB RAM, 2.19467 GB VRAM