Open doopidydoopidy opened 4 years ago
Same issue here. Any solutions?
I found this in the forums:
It means that there are no PNG images in the data_dst folder. If in the data_dst folder there's just "aligned" folder, add the .PNG images who're needed to merge final images in the "merge" folder.
same problem with quick96 merge
edit: i think i solved it. did you delete the extracted images in src or dst folder? probably it needs these files for merging, not only the aligned ones. re-exctract images from videos and try again.
edit2: yep, i solved it by doing the thing i said above
same problem in saehd merge、
I found this in the forums:
It means that there are no PNG images in the data_dst folder. If in the data_dst folder there's just "aligned" folder, add the .PNG images who're needed to merge final images in the "merge" folder.
mean is jpg is not work?
I found this in the forums: It means that there are no PNG images in the data_dst folder. If in the data_dst folder there's just "aligned" folder, add the .PNG images who're needed to merge final images in the "merge" folder.
mean is jpg is not work?
I have no idea since no sane person uses a compressed format in their destination folder. Try with PNG.
Does this issue still persist? If not, would you mind closing it?
Expected behavior
I am trying to merge my files after training the model and receive error No frames to merge in input_dir.
Actual behavior
Initializing models: 100%|###############################################################| 4/4 [00:00<00:00, 5.41it/s] ================== Model Summary ================== == == == Model name: OP_SAEHD == == == == Current iteration: 549949 == == == ==---------------- Model Options ----------------== == == == resolution: 128 == == face_type: wf == == models_opt_on_gpu: True == == archi: df == == ae_dims: 256 == == e_dims: 64 == == d_dims: 64 == == d_mask_dims: 22 == == masked_training: True == == eyes_prio: False == == lr_dropout: False == == random_warp: True == == gan_power: 0.0 == == true_face_power: 0.0 == == face_style_power: 0.0 == == bg_style_power: 0.0 == == ct_mode: none == == clipgrad: False == == pretrain: False == == autobackup_hour: 0 == == write_preview_history: False == == target_iter: 0 == == random_flip: True == == batch_size: 3 == == == ==----------------- Running On ------------------== == == == Device index: 0 == == Name: GeForce RTX 2070 SUPER == == VRAM: 8.00GB == == ==
[y] Use interactive merger? ( y/n ) : y Collecting alignments: 100%|####################################################| 13570/13570 [00:48<00:00, 282.51it/s]
Warning: multiple faces detected. Only one alignment file should refer one source file.
alignment 14720_0.jpg refers to 14720.png alignment 14720_1.jpg refers to 14720.png
It is strongly recommended to process the faces separatelly. Use 'recover original filename' to determine the exact duplicates.
Warning: multiple faces detected. Motion blur will not be used.
No frames to merge in input_dir.
Done. Press any key to continue . . .
SPECS:
i7-6700K 4GHz MSI RTX2070 SUPER 8GB 16BG RAM Windows 10 Python v 3.7.6