Directions: Rewrite each sentence in a Yes/ No question.   She usually calls her mom in the evening.

Question 9 – Part 2 of 3 parts A grid search explores 1000 hyperparameters sets on the validation data in each fold (e.g., layer counts, nodes counts, weight decay). How many neural networks are trained throughout the 10-fold cross validation? Type your answer in the text box.

Directions: Choose the correct words to complete each sentence.   I promise I ______________ with your assignment once I arrive.

Question 7 – Part 3 of 3 parts Assume the actual class of an instance is + class (positive class). Select the output from the softmax function which generates the greatest increase to the cross entropy loss function for this instance.

Question 8 – Part 1 of 3 parts Type your answer in text the box. Consider a convolutional NN with local connectivity and weight sharing for a 32 x 32 pixels (= 1024 pixel) image. Weight sharing includes constant (bias) weights. The 1st hidden layer is generated from an 8 x 8 filter with a stride of 4 over the input. How many nodes are in the 1st hidden layer? Need NOT simplify expressions with numbers.

Question 8 – Part 3 of 3 parts Type your answer in text the box. Again consider a convolutional NN with local connectivity and weight sharing for a 32×32 pixels (= 1024 pixel) image. Weight sharing includes constant (bias) weights. Again the 1st hidden layer is generated from 36 filters, and each is an 8×8 filter with a stride of 4 over the input. A 2nd hidden layer uses 24 filters each of size 4×4 and a stride of 1. Each filter is applied to the depth of nodes in the 1st hidden layer with different weights for each depth. How many weights are estimated between the 1st and 2nd hidden layers (including bias weights)? Need NOT simplify expressions with numbers.

Question 7 – Part 2 of 3 parts If the output from the softmax is yhat = (y1hat, y2hat) = (0.64, 0.36), to what class is the instance assigned?  

Question 9 – Part 1 of 3 parts A dataset has 100,000 instances. Suppose that a classifier is trained with 10-fold cross validation. In each fold, the training set is further partitioned so that 30% of the training set in the fold is used as validation data to select hyperparameters. How many instances are used to evaluate hyperparameters in each fold? Type your answer in the text box.

Match the specific biological route of daily water loss on the left with its correct description or defining characteristic on the right. 

A Burgers circuit drawn around a dislocation normally fails to close when compared with an equivalent circuit in a perfect crystal.