Consider that you are training a classifier for a 10-class classification problem. Each input is represented as a 512-dimensional vector. There are 1000 samples, out of which first 100 will be used for testing. Let Leave-One-Out-Cross-Validation (LOOCV) be used for selection of the classifier model before testing.
Which of the following options is the correct number of validation splits that will be generated?
10
512
900
1000
ANSWER
The correct option is (C): LOOCV generates 900 validation splits.
Official answer: CSource: Official answer key, GATE 2026 DA (IIT Guwahati)
The 100 test samples are held out and must not take part in model selection, so the pool available for LOOCV is samples.
Leave-one-out cross-validation is -fold cross-validation with , where is the size of the pool: each sample is held out once as the validation set and the model is trained on the other samples.
So the number of validation splits equals the pool size: .
The number of classes (10) and the input dimension (512) do not change how many splits LOOCV makes, so options (A) and (B) are distractors.
The count 1000 would include the 100 test samples, which would leak test data into model selection, so (D) is wrong.
Hence 900 splits, each with 899 training samples and 1 validation sample.
KEY IDEA
LOOCV makes one split per sample in the pool used for model selection. The held-out test set is excluded from that pool, so the number of splits is .
WATCH OUT FOR
Answering 1000 by counting every sample, which pulls the test data into model selection. Another slip is to link the split count to the number of classes (10) or the feature dimension (512), which have no bearing on it.
From the GATE 2026 DA paper. Every question of this paper · Download the full paper · All 2 Bias–variance & CV questions, with notes
