openai 的 Clip 模型可以使用 RGBA png 图像吗

Question

大家下午好。我想尝试将 Clip 模型与我的扩散模型合并，但是当我尝试运行代码时，出现错误。 我使用噪声作为输入 (1,4,64,64)，但是当我使用 (1,3,64,64) 时一切正常。

导入剪辑

import open_clip

clip_model, _, preprocess = open_clip.create_model_and_transforms(
    "ViT-B-32", pretrained="openai"
)
clip_model.to(device)

# Transforms to resize and augment an image + normalize to match CLIP's training data
tfms = torchvision.transforms.Compose(
    [
        #torchvision.transforms.RandomResizedCrop(224), 
        torchvision.transforms.RandomAffine(
            5
        ),
        torchvision.transforms.RandomHorizontalFlip(),  
        torchvision.transforms.Normalize([0.5],[0.5]
            mean=(0.48145466, 0.4578275, 0.40821073, 0.4578275),
            std=(0.26862954, 0.26130258, 0.27577711, 0.26130258),
        ),
    ]
)

# And define a loss function that takes an image, embeds it and compares with
# the text features of the prompt
def clip_loss(image, text_features):
    image_features = clip_model.encode_image(
        tfms(image)
    )  # Note: applies the above transforms
    input_normed = torch.nn.functional.normalize(image_features.unsqueeze(1), dim=2)
    embed_normed = torch.nn.functional.normalize(text_features.unsqueeze(0), dim=2)
    dists = (
        input_normed.sub(embed_normed).norm(dim=2).div(2).arcsin().pow(2).mul(2)
    )  # Squared Great Circle Distance
    return dists.mean()

---

代码延续 添加到我的扩散模型中

prompt = "Red Rose (still life), red flower painting"  # @param

# Explore changing this
guidance_scale = 2  # @param
n_cuts = 5 # @param

# More steps -> more time for the guidance to have an effect
scheduler.set_timesteps(50)

# We embed a prompt with CLIP as our target
text = open_clip.tokenize([prompt]).to(device)
with torch.no_grad(), torch.cuda.amp.autocast():
    text_features = clip_model.encode_text(text)


x = torch.randn(4, 4, 64, 64).to(
    device
)  # RAM usage is high, you may want only 1 image at a time

for i, t in tqdm(enumerate(scheduler.timesteps)):

    model_input = scheduler.scale_model_input(x, t)

    # predict the noise residual
    with torch.no_grad():
        noise_pred = image_pipe.unet(model_input, t)["sample"]

    cond_grad = 0

    for cut in range(n_cuts):

        # Set requires grad on x
        x = x.detach().requires_grad_()

        # Get the predicted x0:
        x0 = scheduler.step(noise_pred, t, x).pred_original_sample

        # Calculate loss
        loss = clip_loss(x0, text_features) * guidance_scale

        # Get gradient (scale by n_cuts since we want the average)
        cond_grad -= torch.autograd.grad(loss, x)[0] / n_cuts

    if i % 25 == 0:
        print("Step:", i, ", Guidance loss:", loss.item())

    # Modify x based on this gradient
    alpha_bar = scheduler.alphas_cumprod[i]
    x = (
        x.detach() + cond_grad * alpha_bar.sqrt()
    )  # Note the additional scaling factor here!

    # Now step with scheduler
    x = scheduler.step(noise_pred, t, x).prev_sample


grid = torchvision.utils.make_grid(x.detach(), nrow=4)
im = grid.permute(1, 2, 0).cpu().clip(-1, 1) * 0.5 + 0.5
Image.fromarray(np.array(im * 255).astype(np.uint8))