FuouM-FM_nodes/test.ipynb

In [1]:

import torch
from torchvision import transforms
from PIL import Image

# Function to load and preprocess the image
def load_image_as_tensor(image_path, batch_size=1):
    # Open the image using PIL
    image = Image.open(image_path)
    
    # Define the transformation pipeline
    transform = transforms.Compose([
        transforms.Resize((256, 256)),
        transforms.ToTensor(),  # Convert PIL Image to tensor and scale to [0, 1]
        # transforms.Lambda(lambda x: x.unsqueeze(0).repeat(batch_size, 1, 1, 1))  # Add batch dimension and repeat
    ])
    
    # Apply the transformation
    tensor = transform(image)
    
    return tensor

j:\AI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\utils\hub.py:124: FutureWarning: Using `TRANSFORMERS_CACHE` is deprecated and will be removed in v5 of Transformers. Use `HF_HOME` instead.
  warnings.warn(

In [2]:

image = load_image_as_tensor("examples/006.jpg")

In [3]:

image.shape

Out [3]:

torch.Size([3, 256, 256])

In [4]:

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
scale_max = 2

In [5]:

model_path = "models/anysr/anysr_edsr_500.pth"

In [6]:

model_load = torch.load(model_path)['model']
model_args = model_load['args']['encoder_spec']['args']
pretrain_dict = model_load['sd']

In [7]:

image_tensor = image.squeeze(dim=0).to(device)

In [8]:

image_tensor.shape

Out [8]:

torch.Size([3, 256, 256])

In [9]:

from anysr_module.edsr_anysr import make_edsr


model = make_edsr(scale=2, **model_args)

In [10]:

from anysr_module.anysr_model import replace


model_dict = model.state_dict()
model_dict = replace(pretrain_dict, model_dict)
model.load_state_dict(model_dict, strict=False)

In [11]:

model = model.to(device)

In [12]:

model.scale = 2
model.scale2 = 2

In [14]:

image_tensor.shape

Out [14]:

torch.Size([3, 256, 256])

In [12]:

pred = model(
    image_tensor.unsqueeze(dim=0).cuda(),
)

In [16]:

pred.shape

Out [16]:

torch.Size([1, 256, 256, 256])

In [13]:

pred.shape

Out [13]:

torch.Size([1, 256, 256, 256])

In [14]:

reshaped_tensor = pred.reshape(1, 256, 256, 3, 9, 9)

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
Cell In[14], line 1
----> 1 reshaped_tensor = pred.reshape(1, 256, 256, 3, 9, 9)

RuntimeError: shape '[1, 256, 256, 3, 9, 9]' is invalid for input of size 16777216

In [17]:

pred

Out [17]:

tensor([[[[-0.2023, -0.2465, -0.1883,  ...,  0.0145,  0.0291,  0.0131],
          [-0.2058, -0.3171, -0.2691,  ...,  0.0181,  0.0362,  0.0219],
          [-0.1518, -0.3128, -0.2633,  ...,  0.0039,  0.0277,  0.0250],
          ...,
          [-0.0476,  0.0612,  0.0954,  ..., -0.1948, -0.2188, -0.1279],
          [-0.0426,  0.0376,  0.0445,  ..., -0.1631, -0.1892, -0.0890],
          [ 0.0051,  0.0672,  0.0702,  ..., -0.0010, -0.0048,  0.0314]],

         [[-0.1818, -0.2641, -0.2487,  ..., -0.1725, -0.1746, -0.1650],
          [-0.0194, -0.1230, -0.1304,  ..., -0.2200, -0.2187, -0.1621],
          [-0.0632, -0.1623, -0.1519,  ..., -0.2183, -0.2122, -0.1603],
          ...,
          [-0.1712, -0.1563, -0.1632,  ..., -0.1246, -0.1534, -0.1306],
          [-0.2195, -0.2493, -0.2216,  ..., -0.1596, -0.1834, -0.1382],
          [-0.1708, -0.2002, -0.1785,  ..., -0.1240, -0.1355, -0.1496]],

         [[ 0.0384, -0.1395, -0.1434,  ...,  0.0056, -0.0025, -0.0254],
          [ 0.0819, -0.0770, -0.1296,  ..., -0.0080, -0.0228, -0.0414],
          [ 0.0429, -0.0804, -0.1248,  ..., -0.0178, -0.0256, -0.0443],
          ...,
          [ 0.1172,  0.0857,  0.0598,  ..., -0.0916, -0.1025, -0.0719],
          [ 0.0778,  0.0410,  0.0348,  ..., -0.0644, -0.0767, -0.0587],
          [ 0.0415,  0.0601,  0.0726,  ...,  0.0271,  0.0418,  0.0895]],

         ...,

         [[ 0.0076, -0.1249, -0.1511,  ..., -0.0041, -0.0041, -0.0131],
          [-0.1826, -0.2713, -0.2433,  ...,  0.0897,  0.0770,  0.0489],
          [-0.1696, -0.2405, -0.2122,  ...,  0.0730,  0.0656,  0.0382],
          ...,
          [ 0.0398,  0.0766,  0.1059,  ..., -0.0038, -0.0599, -0.0317],
          [ 0.0230,  0.0949,  0.1537,  ..., -0.0311, -0.0921, -0.0622],
          [ 0.0083,  0.0128,  0.0458,  ..., -0.1552, -0.1739, -0.1277]],

         [[ 0.1756,  0.2339,  0.2451,  ...,  0.1755,  0.1732,  0.1503],
          [ 0.0185, -0.0176,  0.0165,  ...,  0.2317,  0.2341,  0.1946],
          [-0.0086, -0.0330,  0.0083,  ...,  0.2167,  0.2240,  0.1849],
          ...,
          [ 0.1407,  0.1760,  0.2159,  ...,  0.0839,  0.0529,  0.0533],
          [ 0.1521,  0.2473,  0.3145,  ...,  0.0638,  0.0380,  0.0335],
          [ 0.1696,  0.2355,  0.2771,  ...,  0.0139, -0.0056,  0.0147]],

         [[-0.1895, -0.2206, -0.2065,  ..., -0.0522, -0.0691, -0.0413],
          [-0.1096, -0.1073, -0.1033,  ..., -0.1391, -0.1615, -0.1086],
          [-0.1121, -0.1327, -0.1424,  ..., -0.1380, -0.1547, -0.1007],
          ...,
          [ 0.0065, -0.0212, -0.0236,  ..., -0.1827, -0.1751, -0.0828],
          [-0.0189, -0.0623, -0.0892,  ..., -0.1393, -0.1310, -0.0676],
          [-0.0020, -0.0198, -0.0286,  ..., -0.0580, -0.0464, -0.0142]]]],
       device='cuda:0', grad_fn=<AddBackward0>)