I need two things in this one.

1. complete the steps 9 and 10 tables using with math lab.

2. Explain this process how working on the math lab?( explain upper task how doing with words)

6. Bare nuclei 7. Bland chomatin 8. Normal nucleoli 9. Mitoses And 2 output classes as follows: 0. Benign 1. Malignant Steps 1 Open MATLAB Online. Create a folder with the name of IS from New> Folder, as shown below. After creating the IS folder double click on it to make it your current folder. Next, create the folder Workshop5. Download disease_classification.mlx attached to the Workshop plan item in Module 5. Click on the upload icon in MATLAB to upload the file disease_classification.mlx from your PC to Workshop5 folder. Open the Live Script disease_classification.mlx by double clicking on it. Live scripts allow you to view and interact with both code and output and can include formatted text, equations, and images. This table below describes different ways to run your code: First, we will load the dataset. Run the first section and observe the variables in the Workspace window. Data for classification problems are set up for a neural network by organizing the data into two matrices, the input matrix (inputs) and the target matrix (targets). Each ith column of the input matrix will have nine elements representing the shape features of extract cellular nuclei (see attributes above). Each corresponding column of the target matrix will have two elements. Benign cancer diagnosis is represented with a one in the first element, and malignant cancer diagnosis as one in the second element. (All other elements are zero). Next run section 2 in the live script to create a pattern recognition network. Since the neural network starts with random initial weights, the results of this example will differ slightly every time it is run. The random seed is set to avoid this randomness. However this is not necessary for your own applications. Two-layer (i.e. one-hidden-layer) feed forward neural networks can learn any input-output relationship given enough neurons in the hidden layer. Layers which are not output layers are called hidden layers. We will try a single hidden layer of 9 neurons for this example. In general, more difficult problems require more neurons, and perhaps more layers. Simpler problems require fewer neurons.You will see the below window. The input and output have sizes of 0 because the network has not yet been configured to match our input and target data. This will happen when the network is trained. Steps Run section 3 in the live script. This will create a multi-layer perceptron (MLP) with one input layer, one hidden layer having 9 neurons, and one output layer. Here you can set the various parameters for network design and training. We will train the network with the training function "trainscg", which updates weight and bias values according to the scaled conjugate gradient method. Training occurs according to trainscg specified training parameters. Observe the specified number of epochs, learning rate, momentum, early stopping, etc. (Note that if parameters are not specified, then the network trains according to the default values https://au.mathworks.com/help/deeplearning/ref/trainscg.html\#d123e164960) To get the network ready to be trained, the samples are divided into training, validation, and test sets as ratios. The training set is used to teach the network. Training continues as long as the network continues improving on the validation set. The test set provides a completely independent measure of network accuracy. At this point the network is ready to be trained by running section 4. You will see a similar window to the following: At this point the network is ready to be trained by running section 4 . You will see a similar window to the following: Neural Network Training (nntraintool) x Masiral Mathanark Algorithms Dota Disision: Random (dividerand) Training Scaled Conjugate Gradient (trainscg) Performance: Cross-Entropy (crossentropy) Celculations: MEX Dinta After training is completed, you can close the pop-up window. Steps To see how the network's performance improved during training, either click the "Performance" and "Confusion" buttons in the training tool or run section 5 in the live script. Performance is measured in terms of mean squared error, and is shown in a log scale. It rapidly decreased as the network was trained. Performance is shown for each of the training, validation, and test sets. In the performance plot, observe where the network started to overfit (i.e., at which number of epochs does the validation performance stop improving?) and note it down. For example, in the figure below the network starts overfiring after 6 epochs. Due to early stopping, training will stop early and all reported performance metrics will be evaluated at best validation performance. One measure of how well the neural network has fit the data is the confusion plot. Here the confusion matrix is plotted across all samples. The confusion matrix shows the percentages of correct and incorrect classifications. Correct classifications are the green squares on the matrices diagonal. Incorrect classifications form the red squares. If the network has learned to classify properly, the percentages in the red squares should be very small, indicating few misclassifications. If this is not the case then further training, or training a network with more hidden neurons, would be advisable. The neural network performance can be analysed for training, validation, and testing data separately, by running section 6 in the live script. The trained neural network can now be tested with the testing samples. This will give us a sense of how well the network will do when applied to data from the real world. Another measure of how well the neural network has fit data is the receiver operating characteristic plot. This shows how the false positive and true positive rates relate as the thresholding of outputs is varied from 0 to 1 . The farther left and up the line is, the fewer false positives need to be accepted in order to get a high true positive rate. The best classifiers will have a line going from the bottom left corner, to the top left corner, to the top right corner, or close to that. In this section, you will evaluate the effect of changing some training parameters on the performance on the network. First you will need to remove early stopping by changing the following settings in section 3 in the live script: - net.divideParam.valRatio =0; (leave other ratios as before) - net.trainParam.max_fail= 100000000000000000000000000000; (inf) - net.trainParam.min_grad=0; (zero) This will allow the training to continue for the full number of epochs or up until the minimum performance gradient reaches zero. For each combination in the table below, you have to re-train your network and re-evaluate the network's performance (i.e., re-run sections 2, 4, 5, and 6). From the output of section 6 line 57, note down the "Percentage of Correct Testing Classification" and fill the table below: Note that although we are training for 1000 epochs, we could still stop the training early as in the figure below (at 556 epochs) based on the minimum performance gradient (i.e., when it reaches zero). Neural Network Training Pertormance [plotperform], Epoch 1000, Maximum epoch rea..r x Fils Edit View Insart Took Now, go back to section 3 in the live script and divide the data such that: net.divideParam.trainRatio =0.8; net.divideParam.valRatio =0; net.divideParam.testRatio =0.1; Re-train the network and fill the following table with "Percentage of Correct Testing Classification" from section 6 . Note down the optimal parameters from steps 9 and 10. Try changing other parameters such as momentum, learning rate, and the number of hidden layers. Observe how they affect the network's performance. To complete this workshop, follow steps 1-11 above to design, train and test a neural network to classify cancer disease. Include your completed tables (Steps 9 and 10). Include your observations from steps 9, 10, and 11