Free Professional Data Engineer Exam Braindumps (page: 21)

Page 20 of 95

Suppose you have a dataset of images that are each labeled as to whether or not they contain a human face. To create a neural network that recognizes human faces in images using this labeled dataset, what approach would likely be the most effective?

  1. Use K-means Clustering to detect faces in the pixels.
  2. Use feature engineering to add features for eyes, noses, and mouths to the input data.
  3. Use deep learning by creating a neural network with multiple hidden layers to automatically detect features of faces.
  4. Build a neural network with an input layer of pixels, a hidden layer, and an output layer with two categories.

Answer(s): C

Explanation:

Traditional machine learning relies on shallow nets, composed of one input and one output layer, and at most one hidden layer in between. More than three layers (including input and output) qualifies as "deep" learning. So deep is a strictly defined, technical term that means more than one hidden layer.

In deep-learning networks, each layer of nodes trains on a distinct set of features based on the previous layer's output. The further you advance into the neural net, the more complex the features your nodes can recognize, since they aggregate and recombine features from the previous layer.

A neural network with only one hidden layer would be unable to automatically recognize high-level features of faces, such as eyes, because it wouldn't be able to "build" these features using previous hidden layers that detect low-level features, such as lines.

Feature engineering is difficult to perform on raw image data.

K-means Clustering is an unsupervised learning method used to categorize unlabeled data.


Reference:

https://deeplearning4j.org/neuralnet-overview



What are two of the characteristics of using online prediction rather than batch prediction?

  1. It is optimized to handle a high volume of data instances in a job and to run more complex models.
  2. Predictions are returned in the response message.
  3. Predictions are written to output files in a Cloud Storage location that you specify.
  4. It is optimized to minimize the latency of serving predictions.

Answer(s): B,D

Explanation:

Online prediction

.Optimized to minimize the latency of serving predictions.

.Predictions returned in the response message.

Batch prediction

.Optimized to handle a high volume of instances in a job and to run more complex models.

.Predictions written to output files in a Cloud Storage location that you specify.


Reference:

https://cloud.google.com/ml-engine/docs/prediction- overview#online_prediction_versus_batch_prediction



Which of these are examples of a value in a sparse vector? (Select 2 answers.)

  1. [0, 5, 0, 0, 0, 0]
  2. [0, 0, 0, 1, 0, 0, 1]
  3. [0, 1]
  4. [1, 0, 0, 0, 0, 0, 0]

Answer(s): C,D

Explanation:

Categorical features in linear models are typically translated into a sparse vector in which each possible value has a corresponding index or id. For example, if there are only three possible eye colors you can represent 'eye_color' as a length 3 vector: 'brown' would become [1, 0, 0], 'blue' would become [0, 1, 0] and 'green' would become [0, 0, 1]. These vectors are called "sparse" because they may be very long, with many zeros, when the set of possible values is very large (such as all English words).

[0, 0, 0, 1, 0, 0, 1] is not a sparse vector because it has two 1s in it. A sparse vector contains only a single 1.

[0, 5, 0, 0, 0, 0] is not a sparse vector because it has a 5 in it. Sparse vectors only contain 0s and 1s.


Reference:

https://www.tensorflow.org/tutorials/linear#feature_columns_and_transformations



How can you get a neural network to learn about relationships between categories in a categorical feature?

  1. Create a multi-hot column
  2. Create a one-hot column
  3. Create a hash bucket
  4. Create an embedding column

Answer(s): D

Explanation:

There are two problems with one-hot encoding. First, it has high dimensionality, meaning that instead of having just one value, like a continuous feature, it has many values, or dimensions. This makes computation more time-consuming, especially if a feature has a very large number of categories. The second problem is that it doesn't encode any relationships between the categories. They are completely independent from each other, so the network has no way of knowing which ones are similar to each other.

Both of these problems can be solved by representing a categorical feature with an embedding column. The idea is that each category has a smaller vector with, let's say, 5 values in it. But unlike a one-hot vector, the values are not usually 0. The values are weights, similar to the weights that are used for basic features in a neural network. The difference is that each category has a set of weights (5 of them in this case).

You can think of each value in the embedding vector as a feature of the category. So, if two categories are very similar to each other, then their embedding vectors should be very similar too.


Reference:

https://cloudacademy.com/google/introduction-to-google-cloud-machine-learning- engine-course/a-wide-and-deep-model.html






Post your Comments and Discuss Google Professional Data Engineer exam with other Community members:

Exam Discussions & Posts