"update comment"

ltr/lambdaRank.py

@@ -8,6 +8,15 @@ import functools
 
 
 def lambdaRank(input_dim):
+    """
+    lambdaRank is a ListWise Rank Model, input data and label must be sequence
+    https://papers.nips.cc/paper/2971-learning-to-rank-with-nonsmooth-cost-functions.pdf
+    parameters :
+      input_dim, one document's dense feature vector dimension
+
+    dense_vector_sequence format
+    [[f, ...], [f, ...], ...], f is represent for an float or int number
+    """
     label = paddle.layer.data("label",
                               paddle.data_type.dense_vector_sequence(1))
     data = paddle.layer.data("data",
@@ -16,14 +25,24 @@ def lambdaRank(input_dim):
     # hidden layer
     hd1 = paddle.layer.fc(
         input=data,
+        size=128,
+        act=paddle.activation.Tanh(),
+        param_attr=paddle.attr.Param(initial_std=0.01))
+
+    hd2 = paddle.layer.fc(
+        input=hd1,
         size=10,
         act=paddle.activation.Tanh(),
         param_attr=paddle.attr.Param(initial_std=0.01))
     output = paddle.layer.fc(
-        input=hd1,
+        input=hd2,
         size=1,
         act=paddle.activation.Linear(),
         param_attr=paddle.attr.Param(initial_std=0.01))
+
+    # evaluator
+    evaluator = paddle.evaluator.auc(input=output, label=label)
+    # cost layer
     cost = paddle.layer.lambda_cost(
         input=output, score=label, NDCG_num=6, max_sort_size=-1)
     return cost, output
@@ -83,20 +102,23 @@ def lambdaRank_infer(pass_id):
 
     infer_query_id = None
     infer_data = []
-    infer_data_num = 1000
+    infer_data_num = 1
     fill_default_test = functools.partial(
         paddle.dataset.mq2007.test, format="listwise")
     for label, querylist in fill_default_test():
         infer_data.append(querylist)
         if len(infer_data) == infer_data_num:
             break
+
+    # predict score of infer_data document. Re-sort the document base on predict score
+    # in descending order. then we build the ranking documents
     predicitons = paddle.infer(
         output_layer=output, parameters=parameters, input=infer_data)
     for i, score in enumerate(predicitons):
-        print score
+        print i, score
 
 
 if __name__ == '__main__':
     paddle.init(use_gpu=False, trainer_count=4)
-    train_lambdaRank(100)
-    lambdaRank_infer(pass_id=2)
+    train_lambdaRank(2)
+    lambdaRank_infer(pass_id=1)

ltr/ranknet.py

-import os, sys
+import os
+import sys
 import gzip
 import functools
 import paddle.v2 as paddle
@@ -37,7 +38,7 @@ def half_ranknet(name_prefix, input_dim):
 
 def ranknet(input_dim):
     # label layer
-    label = paddle.layer.data("label", paddle.data_type.integer_value(1))
+    label = paddle.layer.data("label", paddle.data_type.dense_vector(1))
 
     # reuse the parameter in half_ranknet
     output_left = half_ranknet("left", input_dim)
@@ -106,10 +107,9 @@ def ranknet_infer(pass_id):
         gzip.open("ranknet_params_%d.tar.gz" % (pass_id - 1)))
 
     # load data of same query and relevance documents, need ranknet to rank these candidates
-    infer_query_id = None
+    infer_query_id = []
     infer_data = []
-    infer_score_list = []
-    infer_data_num = 1000
+    infer_doc_index = []
 
     # convert to mq2007 built-in data format
     # <query_id> <relevance_score> <feature_vector>
@@ -117,17 +117,19 @@ def ranknet_infer(pass_id):
         paddle.dataset.mq2007.test, format="plain_txt")
 
     for query_id, relevance_score, feature_vector in plain_txt_test():
-        if infer_query_id == None:
-            infer_query_id = query_id
-        elif infer_query_id != query_id:
-            break
+        infer_query_id.append(query_id)
         infer_data.append(feature_vector)
-    predicitons = paddle.infer(
+
+    # predict score of infer_data document. Re-sort the document base on predict score
+    # in descending order. then we build the ranking documents
+    scores = paddle.infer(
         output_layer=output, parameters=parameters, input=infer_data)
+    for query_id, score in zip(infer_query_id, scores):
+        print "query_id : ", query_id, " ranknet rank document order : ", score
 
 
 if __name__ == '__main__':
     paddle.init(use_gpu=False, trainer_count=4)
-    pass_num = 10
+    pass_num = 2
     train_ranknet(pass_num)
     ranknet_infer(pass_id=pass_num - 1)