Learn Before
Whether to Add Contract Scans to a Plant Photo Classifier
Case context: You are training a classifier that recognizes common garden plants from smartphone photos. Your dev and test sets contain ordinary outdoor pictures of plants, leaves, and flowers. A vendor offers you a very large collection of scanned legal contracts, which contain only typed text and no plant images.
Question: Using the relevant machine learning guidance, decide whether the contract scans should be added to the training set, and explain what happens to training cost and model capacity if they are included.
Sample answer: The contract scans should be excluded from training. They are far outside the target data distribution and do not provide useful signal for identifying plants. If they were included, training would take more computation and the model would spend some of its capacity learning patterns that are irrelevant to the plant classification task.
Key points:
- Exclude the contract scans from training.
- They are unlike the dev/test examples and add little or no value.
- Including them increases training cost and uses model capacity on irrelevant features.
Rubric: The student should state that the scans should be excluded. They must explain that including them wastes computation and model capacity on features that do not help the target classification task.
0
1
Tags
Machine Learning
Deep Learning
Supervised Learning
Dive into Deep Learning @ D2L
Data Science
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
Why Skip Data That Does Not Match the Evaluation Set?
True or False: If the training, validation, and test sets are drawn from the same distribution, collecting more training examples will always make the model perform better.
According to machine learning strategy, if the dev error curve has _____, adding more training data is unlikely to help you reach your target.
Why might you exclude data that does not add useful information for training?
True or False: More training data always improves validation accuracy.
If the dev error curve has _____, adding more training data is unlikely to move you toward the target performance.
Match each machine-learning concept with the description that fits it best when deciding whether collecting more data is worthwhile.
Order the steps for using a learning curve to judge whether collecting more training data is worthwhile.
In a leaf-disease classifier, why should a large archive of handwritten invoices be left out of training?
True or False: Examining the learning curve can help you avoid spending months collecting more data that later turns out not to improve validation performance.
When compute is limited, examples that add no _____ should be left out of training.
Match each data scenario to the best action.
Order the steps for deciding whether to add a new data source to training.
Why Irrelevant Training Data Should Be Excluded
Whether to Add Contract Scans to a Plant Photo Classifier
Irrelevant Training Data and Model Capacity