A node-based image processing GUI aimed at making chaining image processing tasks easy and customizable. Born as an AI upscaling application, chaiNNer has grown into an extremely flexible and powerful programmatic image processing application.
for example, after manual conversion of femasr to fp16, onnx version runs OK, while Chainner not recognize manually converted fp32 model to fp16:
code for conversion to onnx
python konverzija_u_onnx.py -s 2 -o femasr_x2.onnx
https://pastebin.com/4Z7Hr22s
Chainner executes model properly.
However after conversion fp32 of pytorch version to fp16
python pytorch_fp16.py -s 2 -o femasr_x2fp16.pth
https://pastebin.com/rni0gUU1
fp16 model not recognized in chainner. Could this issue be resolved?
Faster inference is at the 1st place. :)
Thanks
for example, after manual conversion of femasr to fp16, onnx version runs OK, while Chainner not recognize manually converted fp32 model to fp16: code for conversion to onnx python konverzija_u_onnx.py -s 2 -o femasr_x2.onnx https://pastebin.com/4Z7Hr22s Chainner executes model properly. However after conversion fp32 of pytorch version to fp16 python pytorch_fp16.py -s 2 -o femasr_x2fp16.pth https://pastebin.com/rni0gUU1 fp16 model not recognized in chainner. Could this issue be resolved? Faster inference is at the 1st place. :) Thanks