2
$\begingroup$

The number of samples in my dataset range between 3800 and 100,000 per class. Was wondering if my neural network will be more biased towards the classes with a higher number of images. I'm trying on the pretrained resnet152V2 model from Keras.

$\endgroup$
1

1 Answer 1

0
$\begingroup$

That is an empirical question that can could be answered qualitatively with experimentation. You can train one version with unequal number samples and train another version with equal number of samples. The results between these two versions can be compared to see how bias is due to class imbalances.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.