Neural network with MLPRegression Error "y_true and y_pred have different number of output"

Question

After importing libraries in python, I made a simple neural network,

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)
model_nnr = MLPRegressor(
    hidden_layer_sizes = (30,30,30,30,30),
    activation = 'logistic',
    random_state = 0
).fit(X_train, y_train)

pred_nnr = model_nnr.predict(X_test)

print('r2_score:',r2_score(X_test, pred_nnr))
print('Mean Squared Error:',mean_squared_error(X_test, pred_nnr))

But I faced the following error,

ValueError: y_true and y_pred have different number of output (9!=1)

Can you please help me to fix it? I have 9 input column and 1 output. I searched for the error and found this post,

You need to change the number of neurons in the output layer to 1 or add a new output layer which has only 1 neuron.

But still couldn't find out how to fix my issue.

score 1 · Answer 1 · answered Aug 04 '23 at 22:02

You are calling metrics wrong.

print('r2_score:',r2_score(X_test, pred_nnr))
print('Mean Squared Error:',mean_squared_error(X_test, pred_nnr))

should be

print('r2_score:',r2_score(y_test, pred_nnr))
print('Mean Squared Error:',mean_squared_error(y_test, pred_nnr))

as you want to compare predictions to real labels, not real inputs

Neural network with MLPRegression Error "y_true and y_pred have different number of output"

1 Answers1