Consider the CIFAR-10 common benchmark classification problem using TensorFlow: https://www.tensorflow.org/tutorials/images/enn The task is to classify 3232 pixel RGB images in 10 categories: airplane, automobile, bird, cat, deer, dog, frog, horse, ship, and truck. 1) Try to reproduce the results for the model described in the tutorial, which is found in the link given above. 2) Play with the hyperparameters of the model. For example, as a set of parameters (optimizer, initial learning rate, convolutional kernel size), use some of the following combinations: - optimizer: Adam and stochastic gradient descent. - initial learning rate: 0.1,0.01,0.001,0.0001,0.00001. - convolutional kernel size: 3,5,7 or 9 . - number of epochs: 10,15,18 Please record the accuracy for all of the above combinations (for instance: (Adam, 0.1, 3, 10) ) Provide a report with all the accuracies (as individual figures- the ones produced by Tensorflow, as well as a summary Table, where you report the accuracy for each combination). IMPORTANT NOTE: Adam and the other optimizers come with a default learning rate of 0.01. To change it you have to type the following in the appropriate blocks: Import block from tensorflow.keras.optimizers import Adam train block model_compile(optimizer=Adam(learning_rate =0.001), Hint: You may use and edit the source code provided by the tutorial. Alternatively, you may use the Keras high-level API in order to define, train, and evaluate your model..