StevenLOL
commented
6 years ago En, after reading your post here: https://udohsolomon.github.io/machine%20learning/Anomaly-detection/
I think I got your idea:
- The training data here is for unsupervised training
- The test data here is indeed the label for crossval data
- Crossval data+ label are used to estimate the threshold.
Do you mind to explain what is the train/eval/test data in this project? And why they have same length ?