search

ChrisChen1023 / HINT

HINT: High-quality INpainting Transformer with Enhanced Attention and Mask-aware Encoding
MIT License
30 stars 4 forks source link

test a celebhq #11

Closed saeedhaghshenas1402 closed 5 months ago

saeedhaghshenas1402 commented 6 months ago

Hi dear friend I want to use your HINT project to test an image of celeba. I followed the github steps, but what folder should I put the test image in? And what should I do after that? I would be grateful if you could guide me.

ChrisChen1023 commented 6 months ago

Hi dear friend I want to use your HINT project to test an image of celeba. I followed the github steps, but what folder should I put the test image in? And what should I do after that? I would be grateful if you could guide me.

Hi there,

It does not matter where you put the test image, just use the generate the .flist file, basically it is a list for the paths of all testing images/masks. Please use the flist.py to generate the .flist file, and complete your config.yml by changing the path of 'TEST_INPAINT_IMAGE_FLIST' and 'TEST_MASK_FLIST'. That should be working.