# Why does it not need to set test group when using 'rank:pairwise' in xgboost?

I'm new for learning-to-rank. I'm trying to learn the Learning to rank example provided by xgboost. I found that the core code is as follows in rank.py.

train_dmatrix = DMatrix(x_train, y_train)
valid_dmatrix = DMatrix(x_valid, y_valid)
test_dmatrix = DMatrix(x_test)

train_dmatrix.set_group(group_train)
valid_dmatrix.set_group(group_valid)

params = {'objective': 'rank:pairwise', 'eta': 0.1, 'gamma': 1.0,
'min_child_weight': 0.1, 'max_depth': 6}
xgb_model = xgb.train(params, train_dmatrix, num_boost_round=4,
evals=[(valid_dmatrix, 'validation')])
pred = xgb_model.predict(test_dmatrix)

Group data is used in both training and validation sets. But test set prediction does not use group data. I also looked at some explanations to introduce model output such as What is the output of XGboost using 'rank:pairwise'?.

Actually, in Learning to Rank field, we are trying to predict the relative score for each document to a specific query.

My understanding is that if the test set does not have group data, no query is specified. How does the model output the relative score to the specified query?

And I've tried adding test_dmatrix.set_group(group_test). The output results of the two methods are in good agreement like:

[ 1.3535978  -2.9462705   0.86084974 ... -0.23594362  0.712791
-1.633297  ]

So my question as follows:

1. Why does it not need to set test group when using 'rank:pairwise' in xgboost?

2. How can I get label to the specified group query based on the forecasting score results?

Can anybody explain it to me? Thanks in advance.

The output is a score that can be used to rank the samples, and the point in this sort of ranking problem is that you'll only care about ranking samples within the same group (which you think of as being results from a given query).

But that can be safely left to you on the testing set. (Indeed, you might as well only run the prediction for each group separately. You might think about the output in your case as assuming that the test set is all from a single query.) For scoring on the test set, it might matter what the specified groups are, but not for just making predictions.

For training, the group data is needed so the algorithm knows not to calibrate the rankings for intergroup comparisons.

• Can you explain For scoring on the test set, it might matter what the specified groups are, but not for just making predictions in detail? I can understand the group data is needed for training. But if there is no groupid in test set, how can intra-group comparisons within the same group be made to output predictions? – giser_yugang Apr 29 at 2:47
• As I understand it, the actual model, when trained, only produces a score for each sample independently, without regard for which groups they're in. (Indeed, as in your code the group isn't even passed to the prediction. After training, it's just an ordinary GBM.) The interpretation (and hence also scoring the model on the test set) should use these scores to rank the samples only within groups because the model was trained to ignore inter-group interactions. But, thinking again in the context of ranking search results, you'll only predict on a set of pages matching a given query. – Ben Reiniger Apr 30 at 3:05

1. The train/test grouping is a common practice in Machine Learning/Data Science. The objective of this separation is to present some cases (training) so the algorithm can learn the model without memorizing it (overfitting), this means that the model solves the training cases and then gives a solution with the model to the cases in the test dataset. In that way, the solution is general for all the cases. The case in rank:pairwise is the same: You model your training dataset and apply it to the test dataset (of which you don't know the output).

When you have your model applied to the test dataset, you get a solution, which you compare to the solution ($$Y$$) of your test dataset. In that way you have a real solution and a model solution. In Data Science, the comparison of both is the real capacity of your model.

2. How can I get label to the specified group query based on the forecasting score results?

The function

pred = xgb_model.predict(test_dmatrix)

must give you the label you are looking for. What's wrong with this code?

Note: Is a good but not so common practice to verify with train/test grouping if your model doesn't overfit, and after verifying it doesn't overfit to get again a model with all the dataset together.

• Your first answer is too broad. My group is not a grouping of training set validation sets, but a query grouping in learning-to-rank. The second problem is that the output of model give scores like list in question, not label. – giser_yugang Apr 29 at 2:52