import torch.utils.data.distributed
import torchvision.transforms as transforms
import torchvision.datasets as datasets
from torch.autograd import Variable
import pandas as pd
from PIL import Image

classes = ('cat', 'dog')
transform_test = transforms.Compose([
    transforms.Resize((224, 224)),
    transforms.ToTensor(),
    transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])
])

DEVICE = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model = torch.load("model_epoch20_lr0.0001_best_epoch2.pth")
model.eval()
model.to(DEVICE)


predictions = []

for index in range(1, 501):
    img_path = f'datasets/datatest/test/{index}.jpg'  # Assuming file names are 1.jpg, 2.jpg, 3.jpg, etc.
    
    # Load the image, preprocess it, and make a prediction
    img = Image.open(img_path)
    img = transform_test(img)
    img = img.unsqueeze(0).to(DEVICE)
    output = model(img)
    _, pred = torch.max(output.data, 1)
    predictions.append(pred.data.item())
    
    
# Create a DataFrame from the predictions
df = pd.DataFrame(predictions, columns=['prediction'])


# Save the DataFrame to an Excel file
df.to_excel('predictions.xlsx', index=False)