Issue Assitance regarding you work

[Shyam3225]

Hi ,
I liked your work on MultiImagesteganography and I need your help to resolve a error which encountered while I am executing the code

I am getting a Type error while executing the below code of yours
I am also attaching a image please check and help me out.

#1
def train(model,epochs,decoder_criterion,full_model_optimizer,full_model_criterion,learning_rate,training_iterator,valid_iterator,print_every=50):

training_full_model_loss_list = []
decoder_loss_list = []
valid_loss_list = []

for epoch in range(epochs):
    for index,training_dict in enumerate(training_iterator):
        cover_image = training_dict['cover_image']
        cover_image = cover_image.to(device)
        
        secret_image_1 = training_dict['secret_image_1']
        secret_image_1 = secret_image_1.to(device)
        
        secret_image_2 = training_dict['secret_image_2']
        secret_image_2 = secret_image_2.to(device)
        
        secret_image_3 = training_dict['secret_image_3']
        secret_image_3 = secret_image_3.to(device)
        
        full_model_optimizer.zero_grad()
        
        encoder_output = model(cover_image,secret_image_1,secret_image_2,secret_image_3,secret_image_3,'encoder')

        hidden_image,reveal_image_1,reveal_image_2,reveal_image_3 = model(cover_image,
                                                             secret_image_1,
                                                             secret_image_2,
                                                             secret_image_3,secret_image_3,'full')
        
        full_model_loss = full_model_criterion(hidden_image,cover_image,
                         reveal_image_1,secret_image_1,
                         reveal_image_2,secret_image_2,
                         reveal_image_3,secret_image_3,
                        )
        full_model_loss.backward()
        full_model_optimizer.step()
        
        full_model_optimizer.zero_grad()
        reveal_output1, reveal_output2,reveal_output3 = model(cover_image,
                                                             secret_image_1,
                                                             secret_image_2,
                                                             secret_image_3,encoder_output,'decoder')
        decoder_loss = decoder_criterion(reveal_output1, reveal_output2,reveal_output3,secret_image_1,
                                        secret_image_2,secret_image_3)
        
        decoder_loss.backward()
        full_model_optimizer.step()
        
    training_full_model_loss_list.append(full_model_loss)
    decoder_loss_list.append(decoder_loss)
    if epoch % print_every == 0:
        print("Training full model loss at {} epochs is: {}".format(epoch, full_model_loss))
        print("Training decoder loss at {} epochs is: {}".format(epoch, decoder_loss))
    
return model, training_full_model_loss_list,decoder_loss_list

#2
model, training_full_model_loss_list,decoder_loss_list = train(model,EPOCHS,decoder_criterion,full_model_optimizer,full_model_criterion,LEARNING_RATE,train_data_loader,valid_data_loader,50)