Yang Gu comments

Results 29 comments of


                                            Yang Gu

major difference in performance between sync and async methods to download tensor data

Thank you so much @vladmandic ! This is not a very outstanding improvement, but 4ms delay is real and critical for an application that targets 60FPS, or cares about every...

major difference in performance between sync and async methods to download tensor data

I have updated the PR, and it's under final review. @vladmandic May I ask your favor to test it again? I don't want to introduce any regression for your cases....

major difference in performance between sync and async methods to download tensor data

@vladmandic, For the background, I asked @shaoboyan to investigate the diff between data and dataSync (116ms vs. 94ms) even if we use the custom setTimeout as you reported below. using...

major difference in performance between sync and async methods to download tensor data

> as soon as `@tensorflow/tfjs-backend-wasm` is loaded and `USE_SETTIMEOUTCUSTOM` is enabled, there are problems regardless if WASM is used or not as WASM init code happens on module load >...

major difference in performance between sync and async methods to download tensor data

@vladmandic Sorry, I was occupied by some other things. @mattsoulanille, do you have any suggestion on comment 17?

Compare pose-detection prediction data with another GPU backend

@haoyunfeix PTAL

Compare pose-detection prediction data with another GPU backend

@Linchenn If pose-detection will be removed very soon, we don't need this patch. Otherwise, we hope to bring this in temporarily so that our daily test can pass.

Compare pose-detection prediction data with another GPU backend

Talked with Jiajia about new benchmarks, and she will leave some more comments in the spreadsheet. I will close this PR once the new benchmarks are merged.

Remove fromPixelsAsync

@qjia7 @shaoboyan PTAL

Remove fromPixelsAsync

@mattsoulanille thanks!