devalexqt

Very slow

1

To transfer 100M test file it require 30 seconds with 1GBit connection. (quiche-server -> quiche-client)

Is it possible to export model to use in ffmpeg filters like: https://ffmpeg.org/ffmpeg-filters.html#toc-dnn_005fprocessing-1

No async support for minifyCSS and minifyJS ?

4

No async support ?

Converted fp16 or int8 model require up to 10 minutes to startup.

6

Why it's take so long and take almost 30GB of GPU memory? Is it rebuilding model every time then I run it? Can you fix it?

Hi is it possible to export model to "colorize.pb" format what can be used in ffmpeg dnn_processing filter https://ffmpeg.org/ffmpeg-filters.html#toc-dnn_005fprocessing-1 ? similar to this `./ffmpeg -i rain.jpg -vf format=rgb24,dnn_processing=dnn_backend=tensorflow:model=can.pb:input=x:output=y derain.jpg`