The questions will focus on the hyperparameters.
- The number of k clusters is not that important. (Yes | No)
- Mini-batches and batches contain the same amount of data. (Yes | No)
- K-means can run without mini-batches. (Yes | No)
- Must centroids be totally optimized for result acceptance? (Yes | No)
- It does not take long to optimize hyperparameters. (Yes | No)
- It sometimes takes weeks to train a large dataset. (Yes | No)
- AWS SageMaker has a k-means algorithm only. (Yes | No)