Suppose you use Batch Gradient Descent to train a neural network and you plot the training error at every epoch. If you notice that the training error consistently goes up, what is likely going on? How can you fix this?

Write a program that prompts the user to input a string. the program then uses the function substr to remove all the vowels from the string. for example, if str = "there", then after removing all the vowels, str = "thr". after removing all the vowels, output the string. your program must contain a function to remove all the vowels and a function to determine whether a character is a vowel.