I have processes the data in same way as taught in SVM lecture and have used SVM, RandomForest, CNN on data but still i am getting very low accuracy of 43-44% only. I have also tried GridSearch to tune SVM’s Hyperparameters. Please help me figure out in which part i am doing mistake.
Getting very low accuracy
Hey @rk221_b, try to check the error plot for all three classifiers in svm multiclass classification. And see if plots are correct or not. Also remember to normalize your images.
By normalize you mean bring it in ragne of 0-1 by dividing 255.0 …? and for loss in neural net (CNN) its decreasing till 0.5 but after that its fluctuating in this range only and accuracy is also fluctuating in range of 69-70 only
Hey @rk221_b, yes by normalizing it i mean bringing it to range between 0-1. The problem actually needs to be done using svm’s multiclass classification. So first focus on that part, and as i said, plot the error curve for all three different svm classifiers and see if no spikes are there on it. If spikes are there, decrease the learning rate, keep reducing it until curve smoothens.
Hope this resolved your doubt.
Plz mark teh doubt as resolved in my doubts section. 
I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.
On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.