Error on RIFE Pass

[Fllamber]

While interpolating on RIFE pass 2 it encountered an exception

RIFE pass 2 Interpolating 278 images Error completing request

RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 512 but got size 1280 for tensor number 1 in the list.

The problem was encontered while experimenting on higher resolutions with the Hires. fix box ticked

RIFE pass 1
Interpolating 139 images
RIFE pass 2
Interpolating 278 images
Error completing request
Arguments: ('task(7j8zywk2ynhg6x9)', 'hyper detailed, photorealistic,  BioShock Infinite outside sky scene, ((floating city in the sky)), sky, cozy interior lighting, by shinkai makoto studio ghibli, Beaux Arts, Neoclassical, Art Nouveau, long shot blue hour,', 'people, persons, ocean, sea', [], 60, 16, False, False, 1, 1, 14.5, 3569735251.0, -1.0, 0, 0, 0, False, 512, 768, True, 0.55, 2.5, 'Lanczos', 30, 0, 0, 0
, '', '', ['Clip skip: 1'], 7, '\n    <div style="padding: 10px">\n      <div>Estimated VRAM usage: <span style="color: rgb(255.00, 0.00, 204.00)">14226.62 MB / 12288 MB (115.78%)</span></div>\n      <div>(45
    res = list(func(*args, **kwargs))
  File "C:\stable-diffusion-webui\modules\call_queue.py", line 37, in f
    res = func(*args, **kwargs)
  File "C:\stable-diffusion-webui\modules\txt2img.py", line 54, in txt2img
    processed = modules.scripts.scripts_txt2img.run(p, *args)
  File "C:\stable-diffusion-webui\modules\scripts.py", line 441, in run
    processed = script.run(p, *script_args)
  File "C:\stable-diffusion-webui\extensions\seed_travel\scripts\seed_travel.py", line 552, in run
    output = execute(I0, I1, 1)
  File "C:\stable-diffusion-webui\extensions\seed_travel\scripts\seed_travel.py", line 513, in execute
    res.append(rifemodel.inference(I0, I1, (i+1) * 1. / (n+1), scale))
  File "C:\stable-diffusion-webui\extensions\seed_travel\scripts\rife\RIFE_HDv3.py", line 59, in inference
    imgs = torch.cat((img0, img1), 1)
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 512 but got size 1280 for tensor number 1 in the list

This is the second time I lose a generation and it takes so much time even for small seed steps with the RIFE pass, could you also include some way of restoring a video generation? It could be as simple as a file with the images numbers on it, just so that one could redo the video.

[yownas]

There should be two files in the folder where the video is saved.

seed_travel-info-<number>.txt - that will include all settings used

travel-<number>.mp4 - The "un-RIFEed" video. I deliberately save a copy before running RIFE, just to have a backup of sorts if somethings fails.

But apart from that, it shouldn't crash. It looks like it has something to do with upscaling. I'll take a closer look when I get back home from work.

Does it crash just at the start of the second pass?