Is it possible to optionally use Numpy? Cupy is a horrible mess when it comes to compatibility.

[n00mkrad]

It's impossible to run this code on any up-to-date Nvidia machine, as Cupy does not support CUDA 11.3.

Redistributing would also be near-impossible as Cupy only works for one specific CUDA version.

[lisiyao21]

The cupy is used in softsplat. I have a substitution but will make the code slower. I can first update it to make you run the code successfully and find a faster way afterwards...

[n00mkrad]

That would be great as I'm looking into implementing this into a GUI that should run on any recent CUDA version.

[lisiyao21]

Hi, I just update a version without using cupy. To use it, please substitute "model = AnimeInterp" in config file to "model = AnimeInterpNoCupy" and comment the the content about AnimeInterp (import & all) in models/init.py

The substitution seems slightly affect the results but it should not be too much.

[n00mkrad]

Hi, I just update a version without using cupy. To use it, please substitute "model = AnimeInterp" in config file to "model = AnimeInterpNoCupy" and comment the the content about AnimeInterp (import & all) in models/init.py

The substitution seems slightly affect the results but it should not be too much.

Thanks for the dedication, but I still get errors as it's trying to import CuPy either way.

Traceback (most recent call last):
  File "D:\Software\Python38\lib\site-packages\cupy\__init__.py", line 16, in <module>
    from cupy import _core  # NOQA
  File "D:\Software\Python38\lib\site-packages\cupy\_core\__init__.py", line 1, in <module>
    from cupy._core import core  # NOQA
  File "cupy\_core\core.pyx", line 1, in init cupy._core.core
  File "D:\Software\Python38\lib\site-packages\cupy\cuda\__init__.py", line 8, in <module>
    from cupy.cuda import compiler  # NOQA
  File "D:\Software\Python38\lib\site-packages\cupy\cuda\compiler.py", line 11, in <module>
    from cupy.cuda import device
  File "cupy\cuda\device.pyx", line 1, in init cupy.cuda.device
ImportError: DLL load failed while importing runtime: Die angegebene Prozedur wurde nicht gefunden.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "test_anime_sequence_one_by_one.py", line 1, in <module>
    import models
  File "D:\Code\GitHub\AnimeInterp\models\__init__.py", line 1, in <module>
    from .AnimeInterp import AnimeInterp
  File "D:\Code\GitHub\AnimeInterp\models\AnimeInterp.py", line 9, in <module>
    from .softsplat import ModuleSoftsplat as ForwardWarp
  File "D:\Code\GitHub\AnimeInterp\models\softsplat.py", line 7, in <module>
    import cupy
  File "D:\Software\Python38\lib\site-packages\cupy\__init__.py", line 37, in <module>
    raise ImportError(_msg) from e
ImportError: CuPy is not correctly installed.

[routineLife1]

init.py

from .AnimeInterp_no_cupy import AnimeInterpNoCupy

__all__ = [ 'AnimeInterp' ]

[n00mkrad]

utils/correlation.py still calls cupy.

[routineLife1]

utils/correlation.py still calls cupy.

utils.init.py

from utils import config

__all__ = ['config', 'correlation', 'EMA']

[routineLife1]

There is no obvious difference between using cupy and not using cupy

[n00mkrad]

CuPy no longer throws errors.

However, it seems like the paths are based on Linux?

Traceback (most recent call last):
  File "test_anime_sequence_one_by_one.py", line 216, in <module>
    psnrs, ssims, ies, psnr, ssim, psnr_roi, ssim_roi, psnrs_level, ssims_level, folder = validate(config)
  File "test_anime_sequence_one_by_one.py", line 95, in validate
    for validationIndex, validationData in enumerate(validationloader, 0):
  File "D:\Software\Python38\lib\site-packages\torch\utils\data\dataloader.py", line 517, in __next__
    data = self._next_data()
  File "D:\Software\Python38\lib\site-packages\torch\utils\data\dataloader.py", line 1199, in _next_data
    return self._process_data(data)
  File "D:\Software\Python38\lib\site-packages\torch\utils\data\dataloader.py", line 1225, in _process_data
    data.reraise()
  File "D:\Software\Python38\lib\site-packages\torch\_utils.py", line 429, in reraise
    raise self.exc_type(msg)
PermissionError: Caught PermissionError in DataLoader worker process 0.
Original Traceback (most recent call last):
  File "D:\Software\Python38\lib\site-packages\torch\utils\data\_utils\worker.py", line 202, in _worker_loop
    data = fetcher.fetch(index)
  File "D:\Software\Python38\lib\site-packages\torch\utils\data\_utils\fetch.py", line 44, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "D:\Software\Python38\lib\site-packages\torch\utils\data\_utils\fetch.py", line 44, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "E:\AI\AnimeInterp\datas\AniTripletWithSGMFlowTest.py", line 198, in __getitem__
    image = _pil_loader(self.framesPath[index][frameIndex], cropArea=cropArea[frameIndex],  resizeDim=self.resizeSize, frameFlip=randomFrameFlip)
  File "E:\AI\AnimeInterp\datas\AniTripletWithSGMFlowTest.py", line 66, in _pil_loader
    with open(path, 'rb') as f:
PermissionError: [Errno 13] Permission denied: 'E:/AI/AnimeInterp/datasets\\test_2k_540p\\Disney_v4_0_000024_s2'

Seems like it does a double backslash when concatenating paths instead of using a forward slash.

[routineLife1]

I run this code with Windows , maybe you can try to remove calculations like PSNR in test_anime_sequence_one_by_one.py and only get the visual effect in store_path

def save_flow_to_img(flow, des):
        f = flow[0].data.cpu().numpy().transpose([1, 2, 0])
        fcopy = f.copy()
        fcopy[:, :, 0] = f[:, :, 1]
        fcopy[:, :, 1] = f[:, :, 0]
        cf = flow_to_color(-fcopy)
        cv2.imwrite(des + '.jpg', cf)


def validate(config):   
    # preparing datasets & normalization
    normalize1 = TF.Normalize(config.mean, [1.0, 1.0, 1.0])
    normalize2 = TF.Normalize([0, 0, 0], config.std)
    trans = TF.Compose([TF.ToTensor(), normalize1, normalize2, ])

    revmean = [-x for x in config.mean]
    revstd = [1.0 / x for x in config.std]
    revnormalize1 = TF.Normalize([0.0, 0.0, 0.0], revstd)
    revnormalize2 = TF.Normalize(revmean, [1.0, 1.0, 1.0])
    revNormalize = TF.Compose([revnormalize1, revnormalize2])

    revtrans = TF.Compose([revnormalize1, revnormalize2, TF.ToPILImage()])

    testset = datas.AniTripletWithSGMFlowTest(config.testset_root, config.test_flow_root, trans, config.test_size, config.test_crop_size, train=False)
    sampler = torch.utils.data.SequentialSampler(testset)
    validationloader = torch.utils.data.DataLoader(testset, sampler=sampler, batch_size=1, shuffle=False, num_workers=1)
    to_img = TF.ToPILImage()
 
    print(testset)
    sys.stdout.flush()

    # prepare model
    model = getattr(models, config.model)(config.pwc_path).cuda()
    model = nn.DataParallel(model)
    retImg = []

    # load weights
    dict1 = torch.load(config.checkpoint)
    model.load_state_dict(dict1['model_state_dict'], strict=False)

    # prepare others
    store_path = config.store_path

    folders = []

    print('Everything prepared. Ready to test...')  
    sys.stdout.flush()

    #  start testing...
    with torch.no_grad():
        model.eval()
        ii = 0
        for validationIndex, validationData in enumerate(validationloader, 0):
            print('Testing {}/{}-th group...'.format(validationIndex, len(testset)))
            sys.stdout.flush()
            sample, flow,  index, folder = validationData

            frame0 = None
            frame1 = sample[0]
            frame3 = None
            frame2 = sample[-1]

            folders.append(folder[0][0])
            
            # initial SGM flow
            F12i, F21i  = flow

            F12i = F12i.float().cuda() 
            F21i = F21i.float().cuda()

            ITs = [sample[tt] for tt in range(1, 2)]
            I1 = frame1.cuda()
            I2 = frame2.cuda()
            
            if not os.path.exists(config.store_path + '/' + folder[0][0]):
                os.mkdir(config.store_path + '/' + folder[0][0])


            revtrans(I1.cpu()[0]).save(store_path + '/' + folder[0][0] + '/'  + index[0][0] + '.jpg')
            revtrans(I2.cpu()[0]).save(store_path + '/' + folder[-1][0] + '/' +  index[-1][0] + '.jpg')
            for tt in range(config.inter_frames):
                x = config.inter_frames
                t = 1.0/(x+1) * (tt + 1)
                
                outputs = model(I1, I2, F12i, F21i, t)

                It_warp = outputs[0]

                to_img(revNormalize(It_warp.cpu()[0]).clamp(0.0, 1.0)).save(store_path + '/' + folder[1][0] + '/' + index[1][0] + '.png')
                
                save_flow_to_img(outputs[1].cpu(), store_path + '/' + folder[1][0] + '/' + index[1][0] + '_F12')
                save_flow_to_img(outputs[2].cpu(), store_path + '/' + folder[1][0] + '/' + index[1][0] + '_F21')

if __name__ == "__main__":

    # loading configures
    parser = argparse.ArgumentParser()
    parser.add_argument('config')
    args = parser.parse_args()
    config = Config.from_file(args.config)

    if not os.path.exists(config.store_path):
        os.mkdir(config.store_path)

    validate(config)

[n00mkrad]

I made it work, I had something else wrong.

[routineLife1]

It runs like this