From 21f5c9cd34d4cf4809e127fe4afb6846ba4e5dbc Mon Sep 17 00:00:00 2001 From: williamleif Date: Thu, 12 Oct 2017 14:17:40 -0700 Subject: [PATCH] Describing mean pooling. --- README.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index d2ac546..f6474e6 100644 --- a/README.md +++ b/README.md @@ -65,7 +65,8 @@ you can use the `run_walks` function in `graphsage.utils`. The user must also specify a --model, the variants of which are described in detail in the paper: * graphsage_mean -- GraphSage with mean-based aggregator * graphsage_seq -- GraphSage with LSTM-based aggregator -* graphsage_pool -- GraphSage with max-pooling aggregator +* graphsage_maxpool -- GraphSage with max-pooling aggregator (as described in the NIPS 2017 paper) +* graphsage_meanpool -- GraphSage with mean-pooling aggregator (a variant of the pooling aggregator, where the element-wie mean replaces the element-wise max). * gcn -- GraphSage with GCN-based aggregator * n2v -- an implementation of [DeepWalk](https://arxiv.org/abs/1403.6652) (called n2v for short in the code.)