From 8d82cdd6bf61b6894d44c2db87953144939ad0d0 Mon Sep 17 00:00:00 2001
From: Tingyu Wang <tingyuw@nvidia.com>
Date: Mon, 11 Sep 2023 11:03:16 -0400
Subject: [PATCH] clean up docstring

---
 .../cugraph_dgl/nn/conv/relgraphconv.py       | 26 ++++++-------------
 .../cugraph_dgl/nn/conv/sageconv.py           |  4 +--
 2 files changed, 10 insertions(+), 20 deletions(-)

diff --git a/python/cugraph-dgl/cugraph_dgl/nn/conv/relgraphconv.py b/python/cugraph-dgl/cugraph_dgl/nn/conv/relgraphconv.py
index 89e49011cf7..aaecebae0d8 100644
--- a/python/cugraph-dgl/cugraph_dgl/nn/conv/relgraphconv.py
+++ b/python/cugraph-dgl/cugraph_dgl/nn/conv/relgraphconv.py
@@ -10,10 +10,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-"""Torch Module for Relational graph convolution layer using the aggregation
-primitives in cugraph-ops"""
-# pylint: disable=no-member, arguments-differ, invalid-name, too-many-arguments
-from __future__ import annotations
+
 import math
 from typing import Optional
 
@@ -29,13 +26,8 @@
 class RelGraphConv(BaseConv):
     r"""An accelerated relational graph convolution layer from `Modeling
     Relational Data with Graph Convolutional Networks
-    <https://arxiv.org/abs/1703.06103>`__ that leverages the highly-optimized
-    aggregation primitives in cugraph-ops.
-
-    See :class:`dgl.nn.pytorch.conv.RelGraphConv` for mathematical model.
-
-    This module depends on :code:`pylibcugraphops` package, which can be
-    installed via :code:`conda install -c nvidia pylibcugraphops>=23.02`.
+    <https://arxiv.org/abs/1703.06103>`__, with the sparse aggregation
+    accelerated by cugraph-ops.
 
     Parameters
     ----------
@@ -84,7 +76,6 @@ class RelGraphConv(BaseConv):
             [-1.4335, -2.3758],
             [-1.4331, -2.3295]], device='cuda:0', grad_fn=<AddBackward0>)
     """
-    MAX_IN_DEGREE_MFG = 500
 
     def __init__(
         self,
@@ -167,12 +158,11 @@ def forward(
             so any input of other integer types will be casted into int32,
             thus introducing some overhead. Pass in int32 tensors directly
             for best performance.
-        max_in_degree : int, optional
-            Maximum in-degree of destination nodes. It is only effective when
-            :attr:`g` is a :class:`DGLBlock`, i.e., bipartite graph. When
-            :attr:`g` is generated from a neighbor sampler, the value should be
-            set to the corresponding :attr:`fanout`. If not given,
-            :attr:`max_in_degree` will be calculated on-the-fly.
+        max_in_degree : int
+            Maximum in-degree of destination nodes. When :attr:`g` is generated
+            from a neighbor sampler, the value should be set to the corresponding
+            :attr:`fanout`. This option is used to invoke the MFG-variant of
+            cugraph-ops kernel.
 
         Returns
         -------
diff --git a/python/cugraph-dgl/cugraph_dgl/nn/conv/sageconv.py b/python/cugraph-dgl/cugraph_dgl/nn/conv/sageconv.py
index cc91ec6ffe9..fe13ed07dea 100644
--- a/python/cugraph-dgl/cugraph_dgl/nn/conv/sageconv.py
+++ b/python/cugraph-dgl/cugraph_dgl/nn/conv/sageconv.py
@@ -24,8 +24,8 @@
 
 class SAGEConv(BaseConv):
     r"""An accelerated GraphSAGE layer from `Inductive Representation Learning
-    on Large Graphs <https://arxiv.org/pdf/1706.02216.pdf>`__ that leverages the
-    highly-optimized aggregation primitives in cugraph-ops.
+    on Large Graphs <https://arxiv.org/pdf/1706.02216.pdf>`, with the sparse
+    aggregation accelerated by cugraph-ops.
 
     Parameters
     ----------