validation_function.py

from sklearn.metrics import accuracy_score
from sklearn.metrics import cohen_kappa_score
from sklearn.metrics import precision_recall_fscore_support

def validation_prediction(dataloader,model,loss_fn): 
    model.eval()     
    total = 0
    correct = 0
    running_loss = 0
    actual_label_list = []
    predicted_label_list = []

    with torch.no_grad():  
        for x,y in dataloader:
            
            output        = model(x.to(device)) 
            loss          = loss_fn(output, y.to(device)).item()
            running_loss += loss
          
            total        += y.size(0)
            predictions   = output.argmax(dim=1).cpu().detach()
            correct      += (predictions == y.cpu().detach()).sum().item()

            actual_label_list.append(y.cpu().detach().tolist())
            predicted_label_list.append(predictions.tolist())
            
    avg_loss = running_loss/len(dataloader)     
    validation_accuracy = 100*(correct/total)

    print(f'\nValidation Loss = {avg_loss:.6f}',end='\t')
    print(f'Accuracy on Validation set = {100*(correct/total):.6f}% [{correct}/{total}]')

    actual_labels = [item for sublist in actual_label_list  for item in sublist]
    predicted_labels = [item for sublist in predicted_label_list  for item in sublist]

    acc = accuracy_score(actual_labels, predicted_labels)
    p, r, f1, _ = precision_recall_fscore_support(actual_labels,  predicted_labels, average='macro')
    kappa = cohen_kappa_score(actual_labels, predicted_labels)

    print('Acc: ' + str(acc*100)+'%')
    print('Precision: ' + str(p*100)+'%')
    print('Recall: ' + str(r*100)+'%')
    print('F1 score: ' + str(f1*100)+'%')
    print('Kappa score: ' + str(kappa*100)+'%')

    return avg_loss, validation_accuracy

valid_loss, validation_accuracy = validation_prediction(valid_dataloader, model, loss_fn)