I am working on image classification problem where I have to identify whether given image is original one or disguised.During data analysis I found that data is evenly distributed amongst class/target but when I saw the distribution for each of the features they have either skewed or bimodal distribution. When I trained my model with both before transformation and after (box-cox) accuracy does not improve, it is stuck at 50% only. I am not getting where I am going wrong or else does this uneven distribution of features does not affect neural network? Please suggest........


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.