From 4ff92408bda870aac029dee8ee762595d46b592d Mon Sep 17 00:00:00 2001
From: JohanObluda <44701960+JohanObluda@users.noreply.github.com>
Date: Wed, 1 Apr 2020 18:06:24 +0200
Subject: [PATCH] Update README.md

---
 README.md | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/README.md b/README.md
index d0bc484..d9ff5ff 100644
--- a/README.md
+++ b/README.md
@@ -1 +1,4 @@
-# DeepLearningProject
\ No newline at end of file
+This project compares performance of 3 word embeddings for the patent classification task using deep neural network, which are:
+- GLOVE embeddings of 100 dimensions which was trained on 6 billion tokens corpus
+- Uncased BERT-base embeddings (12-layers, 768-hiddens)
+- Domain specific embeddings from https://www.researchgate.net/publication/332088506