1. What is Keras?
a. A deep learning framework
b. A programming language
c. A database management system
d. An operating system
Answer: a. A deep learning framework
2. Keras is an open-source neural network library written in:
a. Python
b. C++
c. Java
d. JavaScript
Answer: a. Python
3. Which backend does Keras primarily support?
a. TensorFlow
b. PyTorch
c. Caffe
d. Theano
Answer: a. TensorFlow
4. Keras was developed by:
a. Facebook
b. Google
c. Microsoft
d. OpenAI
Answer: b. Google
5. What is the main objective of Keras?
a. Ease of use and flexibility
b. Maximum computational efficiency
c. High-level abstraction only
d. Strongly typed programming
Answer: a. Ease of use and flexibility
6. Which of the following is true about Keras?
a. Keras is a standalone deep learning framework.
b. Keras is now part of TensorFlow as its high-level API.
c. Keras is primarily used for natural language processing.
d. Keras is not actively maintained.
Answer: b. Keras is now part of TensorFlow as its high-level API.
7. Which of the following is a key advantage of using Keras?
a. Efficient distributed computing
b. Low-level control over network architecture
c. Prototyping and experimentation
d. Integration with C++
Answer: c. Prototyping and experimentation
8. What is the basic data structure in Keras for building neural networks?
a. Tensor
b. Array
c. List
d. Dictionary
Answer: a. Tensor
9. In Keras, how can you define a simple fully connected neural network?
a. Using the Dense layer
b. Using the Conv2D layer
c. Using the LSTM layer
d. Using the MaxPooling2D layer
Answer: a. Using the Dense layer
10. Which Keras layer is typically used for recurrent neural networks (RNNs)?
a. Dense
b. LSTM
c. Conv2D
d. Flatten
Answer: b. LSTM
11. To compile a Keras model, which parameters do you need to specify?
a. Loss function, optimizer, and metrics
b. Activation function, learning rate, and epochs
c. Input shape, output shape, and batch size
d. Layers, nodes, and weights
Answer: a. Loss function, optimizer, and metrics
12. What is the purpose of the loss function in a Keras model?
a. It defines the metric to optimize during training.
b. It measures the quality of predictions made by the model.
c. It specifies the activation function for the output layer.
d. It initializes the weights of the neural network.
Answer: b. It measures the quality of predictions made by the model.
13. In Keras, which optimizer adjusts the model's weights to minimize the loss?
a. Adam
b. SGD
c. RMSprop
d. All of the above
Answer: d. All of the above
14. Which activation function is commonly used in the output layer for binary classification in Keras?
a. Sigmoid
b. Tanh
c. ReLU
d. Softmax
Answer: a. Sigmoid
15. Which activation function is typically used in hidden layers in Keras?
a. Sigmoid
b. Tanh
c. ReLU
d. Softmax
Answer: c. ReLU
16. How can you prevent overfitting in a Keras model?
a. Adding more layers
b. Increasing the learning rate
c. Using dropout and regularization techniques
d. Using a larger batch size
Answer: c. Using dropout and regularization techniques
17. What is data augmentation in the context of Keras?
a. Generating new data by perturbing existing data samples
b. Cleaning and preprocessing the dataset
c. Reducing the dimensions of the input data
d. Scaling the data to a fixed range
Answer: a. Generating new data by perturbing existing data samples
18. How can you save a Keras model to a file for future use?
a. Using the model.save() method
b. Using the model.serialize() method
c. Using the model.export() method
d. Using the model.save_model() method
Answer: a. Using the model.save() method
19. How can you load a saved Keras model from a file?
a. Using the model.load() method
b. Using the load_model() function from keras.models
c. Using the model.deserialize() method
d. Using the load() function from keras.models
Answer: b. Using the load_model() function from keras.models
20. In Keras, what is a callback?
a. A function used to compute loss
b. A function used to compile a model
c. A function used to monitor and customize model training
d. A function used to initialize weights
Answer: c. A function used to monitor and customize model training
21. Which Keras callback can be used to save the model's weights during training?
a. EarlyStopping
b. ModelCheckpoint
c. ReduceLROnPlateau
d. LearningRateScheduler
Answer: b. ModelCheckpoint
22. Which Keras callback can be used to stop training early if a monitored metric has stopped improving?
a. EarlyStopping
b. ModelCheckpoint
c. ReduceLROnPlateau
d. LearningRateScheduler
Answer: a. EarlyStopping
23. In Keras, how can you implement a learning rate schedule?
a. By using the LearningRateScheduler callback
b. By specifying the learning rate in the optimizer
c. By adding more layers to the model
d. By using the ReduceLROnPlateau callback
Answer: a. By using the LearningRateScheduler callback
24. Which of the following statements is true regarding transfer learning in Keras?
a. Transfer learning is the process of training a model from scratch.
b. Transfer learning involves using pre-trained models as a starting point and fine-tuning them for a specific task.
c. Transfer learning is only applicable to image data.
d. Transfer learning is not supported in Keras.
Answer: b. Transfer learning involves using pre-trained models as a starting point and fine-tuning them for a specific task.
25. Which Keras layer is commonly used for transfer learning?
a. Dense
b. Flatten
c. Conv2D
d. LSTM
Answer: b. Flatten
26. In Keras, what does the Flatten layer do?
a. Reduces the dimensionality of the input data
b. Reshapes the input data into a 1D array
c. Applies element-wise activation to the input data
d. Aggregates the input data
Answer: b. Reshapes the input data into a 1D array
27. Which Keras layer is commonly used for image recognition tasks?
a. Dense
b. Flatten
c. Conv2D
d. LSTM
Answer: c. Conv2D
28. What is the purpose of the MaxPooling2D layer in Keras?
a. Reduces the spatial dimensions of the input volume
b. Increases the spatial dimensions of the input volume
c. Applies element-wise activation to the input data
d. Adds noise to the input data
Answer: a. Reduces the spatial dimensions of the input volume
29. Which Keras layer is used for handling sequential data like time series or natural language?
a. Dense
b. Flatten
c. LSTM
d. Conv2D
Answer: c. LSTM
30. In Keras, how can you set the initial weights for a layer?
a. Using the initial_weights parameter in the layer constructor
b. Using the set_weights method after creating the model
c. Weights are automatically initialized in Keras
d. Using the weights parameter in the layer constructor
Answer: d. Using the weights parameter in the layer constructor
Top comments (0)