1. What do Convolutional Neural Networks (CNNs) use to extract numeric feature maps from images?
A) Filters
B) Tokens
C) Embeddings
D) Layers
Explanation: CNNs use filters to extract numeric feature maps from images.
2. What is the output layer of a CNN typically used to produce in an image classification scenario?
A) Probability values for each possible class
B) Feature maps
C) Vector embeddings
D) Tokens
Explanation: The output layer of a CNN uses a softmax or similar function to produce a result that contains a probability value for each possible class.
3. During the training process of a CNN, what is used to calculate the loss in the model?
A) Random weights
B) Initial feature maps
C) Difference between predicted and actual class scores
D) Filter kernels
Explanation: During training, the output probabilities are compared to the actual class label, and the difference between the predicted and actual class scores is used to calculate the loss in the model.
4. What drives most advances in computer vision over the decades?
A) Improvements in CNN-based models
B) Enhancements in data labeling
C) Developments in hardware
D) Growth of big data
Explanation: Most advances in computer vision over the decades have been driven by improvements in CNN-based models.
5. What type of neural network architecture is primarily used in natural language processing (NLP)?
A) Convolutional Neural Network (CNN)
B) Transformer
C) Recurrent Neural Network (RNN)
D) Multi-layer Perceptron (MLP)
Explanation: In natural language processing (NLP), a transformer is primarily used as the neural network architecture.
6. What is an embedding in the context of transformers?
A) A layer in the neural network
B) A type of filter
C) An array of numeric values representing language tokens
D) A classification label
Explanation: An embedding in the context of transformers is an array of numeric values representing language tokens.
7. How are tokens that are commonly used in the same context encoded in transformers?
A) Closer together dimensionally
B) With higher weight values
C) Using filter kernels
D) Randomly
Explanation: Tokens that are commonly used in the same context are encoded closer together dimensionally in transformers.
8. What do multi-modal models combine to create a model that encapsulates relationships between natural language and images?
A) Image filters and class labels
B) CNNs and RNNs
C) Image encoders and text embeddings
D) Feature maps and tokens
Explanation: Multi-modal models combine image encoders and text embeddings to create a model that encapsulates relationships between natural language and images.
9. What is the Microsoft Florence model an example of?
A) A foundation model
B) A CNN model
C) A tokenization model
D) A recurrent model
Explanation: The Microsoft Florence model is an example of a foundation model.
10. What kind of tasks can adaptive models built on the Florence foundation model perform?
A) Text summarization
B) Image classification and object detection
C) Speech recognition
D) Language translation
Explanation: Adaptive models built on the Florence foundation model can perform tasks such as image classification, object detection, captioning, and tagging.
Your score
Total Questions Attempted: 0
Correct Answers: 0
Wrong Answers: 0
Percentage: 0%
0 Comments