NLP Task in TensorFlow

Write a function that implements sigmoid, a function that implements softmax, and in a loop calculate the forward and backward pass, update, check stopping conditions. 1. In the attached Matlab code, enter your ID number in the rng command. The code will generate a training set for a FF neural network. In addition, the code will specify for each student how many neurons will be in the hidden layer. In the submission file, paste the code after filling, paste an image of the training set and provide the number of neurons in the hidden layer. 2. Answer the questions: 1. What are the parameters to train? 2. How many parameters in total does the trainer have? 3. What are the update rules for each parameter (it is also possible to display a joint update rule for vectors or matrices)? 3. Write code to train the neural network. The activation function for neurons in the hidden layer is sigmoid. The activation function of the neurons in the output layer is softmax, and the price function is: Draw a graph of the variability of L according to the number of iterations and attach it to the submission file. Decide (and specify in the submission file) a reasonable initialization for the parameters, a reasonable step size, and a reasonable stopping condition.