CPU Tasks LoadModel 827.449 milliseconds RunComplete 4.95485 seconds Run 4.90459 seconds Callbacks 343.6 microseconds, 4 calls, 85.9 microseconds average Spectrogram 12.0208 milliseconds, 3 calls, 4.00693 milliseconds average Sample 3.798 milliseconds, 28 calls, 135.643 microseconds average Encode 3.78211 seconds Decode 1.12187 seconds DecodeStep 1.11805 seconds, 28 calls, 39.9304 milliseconds average GPU Tasks LoadModel 429.525 milliseconds Run 4.86319 seconds Encode 3.82894 seconds EncodeLayer 3.23846 seconds, 24 calls, 134.936 milliseconds average Decode 1.03424 seconds DecodeStep 1.03083 seconds, 28 calls, 36.8153 milliseconds average DecodeLayer 934.97 milliseconds, 672 calls, 1.39132 milliseconds average Compute Shaders mulMatTiledEx 2.59201 seconds, 240 calls, 10.8 milliseconds average mulMatTiled 733.345 milliseconds, 289 calls, 2.53753 milliseconds average mulMatByRowTiled 492.505 milliseconds, 6507 calls, 75.6884 microseconds average mulMatByRowTiledEx 226.315 milliseconds, 1296 calls, 174.626 microseconds average softMaxFixed 169.603 milliseconds, 696 calls, 243.683 microseconds average convolutionMain2Fixed 130.131 milliseconds matReshapePanels 85.7723 milliseconds, 145 calls, 591.533 microseconds average addRepeatGelu 52.8833 milliseconds, 698 calls, 75.764 microseconds average copyConvert 49.8477 milliseconds, 1440 calls, 34.6165 microseconds average scaleInPlace 47.7803 milliseconds, 696 calls, 68.6499 microseconds average normFixed 44.0434 milliseconds, 2093 calls, 21.0432 microseconds average fmaRepeat1 38.6945 milliseconds, 2093 calls, 18.4876 microseconds average copyTranspose 36.6512 milliseconds, 1392 calls, 26.3299 microseconds average addRepeatEx 33.6887 milliseconds, 2064 calls, 16.322 microseconds average addRepeat 32.9016 milliseconds, 744 calls, 44.2226 microseconds average convolutionMain 32.8426 milliseconds softMaxLong 19.8753 milliseconds, 28 calls, 709.832 microseconds average addRepeatScale 15.8724 milliseconds, 1344 calls, 11.8098 microseconds average softMax 5.5277 milliseconds, 672 calls, 8.22574 microseconds average diagMaskInf 5.1549 milliseconds, 672 calls, 7.67098 microseconds average convolutionPrep2 3.9464 milliseconds, 2 calls, 1.9732 milliseconds average convolutionPrep1 3.4569 milliseconds, 2 calls, 1.72845 milliseconds average add 722.7 microseconds addRows 80 microseconds, 28 calls, 2.85714 microseconds average Memory Usage Model 877.966 KB RAM, 1.42785 GB VRAM Context 1.9831 MB RAM, 771.235 MB VRAM Total 2.84049 MB RAM, 2.18101 GB VRAM