From 32c9f1bbf89ea71ac4b216b6179a6fb82ee5375a Mon Sep 17 00:00:00 2001
From: Victor Zimmermann <zimmermann@cl.uni-heidelberg.de>
Date: Wed, 21 Mar 2018 18:25:32 +0100
Subject: [PATCH] Added maximum context size option.
---
src/absinth.py | 3 ++-
src/config.py | 1 +
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/absinth.py b/src/absinth.py
index d6d22ae..5c16425 100644
--- a/src/absinth.py
+++ b/src/absinth.py
@@ -206,6 +206,7 @@ def process_file(context_list: list, target_string: str,
stopword_list = config.stop_words
allowed_tag_list = config.allowed_tags
min_context_size = config.min_context_size
+ max_context_size = config.max_context_size
try:
@@ -241,7 +242,7 @@ def process_file(context_list: list, target_string: str,
context_size = len(token_set)
- if context_size >= min_context_size:
+ if context_size >= min_context_size and context_size <= max_context_size:
for token in token_set:
if token in node_freq_dict:
diff --git a/src/config.py b/src/config.py
index 01a0a9e..a823503 100644
--- a/src/config.py
+++ b/src/config.py
@@ -41,6 +41,7 @@ max_edges = 2000000
Choose the minimum context size.
'''
min_context_size = 4
+max_context_size = 20
'''
Choose filters for building the graph.
--
GitLab