From bb73005da1c1a3a3c41983e1353dba638407363d Mon Sep 17 00:00:00 2001
From: joriscram <joris.cramwinckel@gmail.com>
Date: Thu, 8 Mar 2018 15:19:46 +0100
Subject: [PATCH] implemented review comments - also changed the readme with
 the instruction to install a spacy nlp model, this way you do not need the
 assets folder.

---
 README.md   | 3 +++
 bin/main.py | 6 +++---
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/README.md b/README.md
index 34b40e2..9b02443 100644
--- a/README.md
+++ b/README.md
@@ -13,6 +13,9 @@ conda env create -f environment.yml
 # Activate Python virtual environment
 source activate resume
 
+#Retrieve language model from spacy
+python -m spacy download en
+
 # Run code (with default configurations)
 cd bin/
 python main.py
diff --git a/bin/main.py b/bin/main.py
index 82bd9ee..3216811 100644
--- a/bin/main.py
+++ b/bin/main.py
@@ -12,8 +12,8 @@
 
 import lib
 import field_extraction
+import spacy
 
-import en_core_web_sm
 
 def main():
     """
@@ -27,7 +27,7 @@ def main():
     observations = extract()
 
     # Spacy: Spacy NLP
-    nlp = en_core_web_sm.load()
+    nlp = spacy.load('en')
 
     # Transform data to have appropriate fields
     observations, nlp = transform(observations, nlp)
@@ -42,7 +42,7 @@ def text_extract_utf8(f):
     try:
         return unicode(textract.process(f), "utf-8")
     except UnicodeDecodeError, e:
-        return e
+        return ''
 
 def extract():
     logging.info('Begin extract')