Learn Before
Automating Short Descriptions for Product Photos
Case context: A retail website wants software that can generate a short description for each product photo uploaded by sellers.
Question: Using the end-to-end image captioning idea, what should the neural network receive as input, and what should it produce as output?
Sample answer: The network should take the product photo as its direct input and generate the text description as its direct output.
Key points:
- Treat the photo as the input x.
- Treat the description as the output y.
- Connect the image directly to the text output in an end-to-end design.
Rubric: The learner must identify this as an image-to-text task and state that the photo is the input (x) and the description is the direct output (y).
0
1
Tags
Python Programming Language
Data Science
Machine Learning
Deep Learning
Supervised Learning
Dive into Deep Learning @ D2L
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
What is the direct output of an end-to-end image captioning network?
A direct image-to-caption system can take a photo as input and produce a caption as output without a separate hand-built intermediate stage.
End-to-End Image Captioning Produces a Direct Description
Match each symbol or term to its role in a direct image-to-text captioning model.
Put the stages of a captioning model’s forward pass in order.
Which statement best describes an end-to-end email spam detector?
Input and output roles in a captioning model
A speech-to-text system that produces full sentences directly is an example of directly learning _____ outputs.
Match each captioning concept to its description.
Order the steps for recognizing a direct mapping from raw input to a detailed output
Explaining End-to-End Image Captioning
Automating Short Descriptions for Product Photos
Inputs and Targets in a Captioning Model