[FIXED] .eq() method is not giving same result as [ == ]

Issue

I am having a hard time understanding why the results are not the same for following code.

I am trying to find the accuracy of a model but the first item gives a result of tensor(66.), and second item gives a result of tensor(105).

(y_test[y_test==y_predicted_cls].sum(), y_predicted_cls.eq(y_test).sum())

Both of the tensors (y_test and y_predicted_cls) have the same data type, torch.float32. Output:

(tensor(66.), tensor(105))

I thought that first one would be equal to the second one.

Solution

The statement y_test[y_test==y_predicted_cls].sum() gives the sum for the y_test list/array while, y_predicted_cls.eq(y_test).sum() gives the sum for y_predicted_cls, and in the first case, if both the arrays are same, it yields:

y_test[1].sum()

else :

y_test[0].sum()

Answered By - Roshin Raphel