Learn Before
Model selection for a massive labeled dataset
Case context: Your team has aggregated and cleaned a massive dataset for a new classification task. The final dataset contains exactly one million labeled training examples.
Question: Based on the guidance from Machine Learning Yearning, what type of model should you decide to favor for this project, and how does the dataset size influence this decision?
Sample answer: You should favor a neural network. Machine Learning Yearning explicitly states that with 1 million examples, a neural network is favored because large neural networks benefit significantly from huge amounts of data.
Key points:
- Recommend using a neural network.
- Identify that 1 million examples is the threshold mentioned to favor this model.
- Note that neural networks benefit from huge data.
Rubric: The answer must recommend a neural network and justify the decision by citing the dataset size of one million examples.
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
Choosing a model with one million examples
Neural networks and one million training examples
The preferred model with _____ training examples
Matching dataset concepts with model selection
Reasoning process for selecting a neural network
Analyzing the need for neural networks with massive data
Model selection for a massive labeled dataset
Preferred model for one million examples
Data condition for favoring a neural network
Insufficiency of one million examples for neural networks