Accuracy, Precision, Recall and F1

def score(targets, predictions):
	tp, tn, fp, fn = 0, 0, 0, 0
	for target, prediction in zip(targets, predictions):
		if target == 1:
			if prediction ==1:
				tp += 1
			else:
				fn += 1
		else:
			if prediction ==1:
				fp += 1
			else:
				tn += 1
	a = (tp+tn) / (tp+tn +fp +fn)
	r = tp / (tp + fn)
	p = tp / (tp + fp)
	f1 = 2 * (p*r)/(p+r)
	return a, r, p, f1

import torch # since I use PyTorch as my DL library!
def scorepro(targets, predictions):
    total = len(targets)
    true_positives = sum(torch.multiply(targets, predictions)) # tp
    possible_positives= sum(targets) # tp + fn
    predicted_positives = sum(predictions) # tp + fp
    
    acc = sum(targets==predictions) / total
    recall = true_positives / possible_positives
    precision = true_positives / predicted_positives

    f1 =  2*((precision*recall)/(precision+recall))
    return acc.item(), recall.item(), precision.item(), f1.item()

targ = [0, 1, 0, 1]
pred = [0, 1, 1, 0]

targt = torch.Tensor(targ)
predt = torch.Tensor(pred)

score(targt, predt) # (0.5, 0.5, 0.5, 0.5)
scorepro(targt, predt) # (0.5, 0.5, 0.5, 0.5)