Search code examples
pythonarraysnumpylabelprediction

Comparing NumPy Arrays (predicted vs actual labels)

I am having issues comparing 2 numPy arrays against each other. I am currently doing a classification task and have classified data into either 1's or 0's. I want to compare the predictions against the actual label.

enter image description here

attached is an image of the two arrays. I cannot seem to find a way .

though not ideal, I have tried the following:

for i in range(len(yTrain)):
    yTrain==yPred

I was hoping that it would return a bunch of true and false values which I could manually tally up but that didn't work.

I would like an output which counts the number of inaccuracies. e.g. out of 200 datapoints, 45 were misclassified or something along those lines

Solution

  • If you want the count of where yTrain == yPred, you can use np.count_nonzero:

    matches = np.count_nonzero(yTrain == yPred)
mismatches = yTrain.size - matches