DLV-2 Semantic Segmentation Pre-Quiz
SHALA2020.GITHUB.IO
Email *
How many trainable parameters (weights and biases) are there in a convolutional layer with 4 filters of size 5x5 and stride 1 that operates upon a colored (3 channel) input image of size 200x200 without zero-padding? Please enter your answers (for all the questions) in digits without any comma or space.
1 point
What is the number of outputs computed by a convolutional layer with 4 filters of size 5x5 and stride 2 that operates upon a colored (3 channel) input image of size 199x199 without zero-padding? If the output is of size W x H x C, then multiply the three numbers W, H, and C to obtain the final answer.
1 point
How many trainable parameters are there in a max-pooling layer of size 2x2 and stride 2 that operates upon a feature map of size 200x200 with 16 channels?
1 point
How many trainable parameters (weights and biases) are there in a fully connected layer with 100 neurons that takes as input a flattened convolutional feature map of size 5x5 with 200 channels?
1 point
Submit
Clear form
Never submit passwords through Google Forms.
This content is neither created nor endorsed by Google. Report Abuse - Terms of Service - Privacy Policy
Google Forms