2. Hinge Loss Bookmark this page In this project you will be implementing linear classifiers beginning with the Perceptron algorithm. You will begin by writing your loss function, a hinge-loss function. For this function you are given the parameters of your model and Do. Additionally, you are given a feature matrix in which the rows are feature vectors and the columns are individual features, and a vector of labels representing the actual sentiment of the corresponding feature vector. Hinge Loss on One Data Sample 1 point possible (graded) First, implement the basic hinge loss calculation on a single data-point. Instead of the entire feature matrix, you are given one row, representing the feature vector of a single data sample, and its label of +1 or -1 representing the ground truth sentiment of the data sample. Reminder: You can implement this function locally first, and run python test.py in your sentiment analysis directory to validate basic functionality before checking against the online grader here. Available Functions: You have access to the NumPy python library as np; No need to import anything. 1 def hinge_1093_single (feature_vector, label, theta, theta_0): Finds the hinge loss on a single data point given specific classification parameters. WOOWN Args : feature_vector - A numpy array describing the given data point. label - Areal valued number, the correct classification of the data point. theta - A numpy array describing the linear classifier. theta_0 - A real valued number representing the offset parameter. un Returns: A real number representing the hinge loss associated with the given data point and parameters. Press ESC then TAB or click outside of the code editor to exit Unanswered Submit You have used 0 of 25 attempts Save Reset