I am trying to build a normal classification model using SVM on Cifar-10 dataset and have applied PCA also but when i am using GridSearch to tune hyper-parameters then its taking too much time and is not finishing. I have used Google Colab with GPU but its not working. What should i do …??
GridSearch and RandomizeSearch taking too long
Hey @rk221_b, svm itself requires very long time for large datasets, and using grid_search on them makes it more worse in terms of training time requirements. Its disappointing but there is no way out for this. You need to use deep learning if you want to do it faster.
Hope this cleared your doubt. 