documentation

pyterrier_dr/prf.py

@@ -14,8 +14,9 @@ class VectorPrf(pt.Transformer):
      - beta: weight of doc_vec
      - k: number of pseudo-relevant feedback documents
 
-    Expected Input: ['qid', 'query_vec', 'docno', 'doc_vec']
-    Output: ['qid', 'query_vec']
+    Expected Input Columns: ``['qid', 'query_vec', 'docno', 'doc_vec']``
+
+    Output Columns: ``['qid', 'query_vec']`` (Any other query columns from the input are also pulled included in the output.)
 
     Example::
     
@@ -56,6 +57,7 @@ def __init__(self,
 
     @pta.transform.by_query(add_ranks=False)
     def transform(self, inp: pd.DataFrame) -> pd.DataFrame:
+        """Performs Vector PRF on the input dataframe."""
         pta.validate.result_frame(inp, extra_columns=['query_vec', 'doc_vec'])
 
         query_cols = [col for col in inp.columns if col.startswith('q') and col != 'query_vec']
@@ -79,8 +81,9 @@ class AveragePrf(pt.Transformer):
     Arguments:
      - k: number of pseudo-relevant feedback documents
 
-    Expected Input: ['qid', 'query_vec', 'docno', 'doc_vec']
-    Output: ['qid', 'query_vec']
+    Expected Input Columns: ``['qid', 'query_vec', 'docno', 'doc_vec']``
+
+    Output Columns: ``['qid', 'query_vec']`` (Any other query columns from the input are also pulled included in the output.)
 
     Example::
     
@@ -117,6 +120,7 @@ def __init__(self,
 
     @pta.transform.by_query(add_ranks=False)
     def transform(self, inp: pd.DataFrame) -> pd.DataFrame:
+        """Performs Average PRF on the input dataframe."""
         pta.validate.result_frame(inp, extra_columns=['query_vec', 'doc_vec'])
 
         query_cols = [col for col in inp.columns if col.startswith('q') and col != 'query_vec']

pyterrier_dr/pt_docs/index.rst

@@ -34,3 +34,11 @@ Import ``pyterrier_dr``, load a pre-built index and model, and retrieve:
     997  82.551933  10064  10063   997   1  chemical reactions
     998  82.546890   4417   4416   998   1  chemical reactions
     999  82.545776   7120   7119   999   1  chemical reactions
+
+
+.. rubric:: Table of Contents
+
+.. toctree::
+    :maxdepth: 1
+
+    prf

pyterrier_dr/pt_docs/prf.rst

@@ -0,0 +1,16 @@
+Pseudo Relevance Feedback (PRF)
+===============================
+
+Dense Pseudo Relevance Feedback (PRF) is a technique to improve the performance of a retrieval system by expanding the
+original query vector with the vectors from the top-ranked documents. The idea is that the top-ranked documents.
+
+PyTerrier-DR provides two dense PRF implementations: :class:`pyterrier_dr.AveragePrf` and :class:`pyterrier_dr.VectorPrf`.
+
+API Documentation
+-----------------
+
+.. autoclass:: pyterrier_dr.AveragePrf
+    :members:
+
+.. autoclass:: pyterrier_dr.VectorPrf
+    :members: