RuntimeError: Input type (c10::Half) and bias type (float) should be the same

[AmineDjeghri]

When running the following code without using model.to(device), it works. But when using the device (which is a GPU), i get the following error : RuntimeError: Input type (c10::Half) and bias type (float) should be the same

Here is the code

import torch
from PIL import Image
import matplotlib.pyplot as plt
from donut import DonutModel
import torch
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

def demo_process(input_img):
    global pretrained_model, task_prompt, task_name
    # input_img = Image.fromarray(input_img)
    output = pretrained_model.inference(image=input_img, prompt=task_prompt)["predictions"][0]
    return output

task_prompt = f"<s_cord-v2>"

image = Image.open("data/sample_image_cord_test_receipt_00004.png")

plt.imshow(image)
plt.show()

pretrained_model = DonutModel.from_pretrained("naver-clova-ix/donut-base-finetuned-cord-v2")
pretrained_model.to(device) 
pretrained_model.eval()

[dreamlychina]

same +1