update stopwords to have widely used slang that I don't need to analyze

Author: freesoft

README.md

@@ -63,12 +63,13 @@ Since the code has tiny TwitchTV chatbot integrated(`chatbot.py`), you can use t
 Simply type
 
 ```
-python3 chatbot.py Moira <client id> <oauth2 acccess token> <channel name>
+python chatbot.py <username> <client id> <oauth2 acccess token> <channel name>
 ```
 and you'll see that the program is collecting the live chat log from the TwhtchTV channel, and determine if chat is toxic or not.
 
 Parameter explanation:
 
+* `<username>` : Your username on Twitch. 
 * `<client id>` : visit https://glass.twitch.tv/ and login with your own Twitch account. Once you create new app, you'll be able to get Client ID on Dashboard -> App section. <br/>
 * `<oauth2 access token>` : visit [here](https://twitchapps.com/tmi/#access_token=flwh72scl6503e6bs2xnwl6g6l5jeu&scope=chat%3Aread+chat%3Aedit+channel%3Amoderate+chat_login&token_type=bearer) and click "Connec with Twitch", and use it for `<oauth2 access token>`.
 * `<channel name>` : TwitchTV channel name you'd like to deploy Moira. Use the channel name you can check from web browser's url, which is generally all lowercase regardless of what you can see on twitch user's dashboard.

chatbot.py

@@ -3,7 +3,7 @@
 import requests
 
 import constant
-from sentiment_twitch import ToxicityClassifier
+from detox_engine import ToxicityClassifier
 
 from joblib import dump, load
 
@@ -98,6 +98,12 @@ def do_command(self, e, cmd):
             print("Did not understand command:" + cmd)
 
 def main():
+
+    if not sys.version_info[:1] == (3,):
+        print(sys.version_info[:1] )
+        sys.stderr.write("Python version 3 is required.\n")
+        exit(1)
+
     if len(sys.argv) != 5:
         print("Usage: twitchbot <username> <client id> <token> <channel>")
         sys.exit(1)

detox_engine.py

@@ -29,7 +29,7 @@
 # 
 # !!!! IMPORTANT !!!
 #
-# Which means if you ever chagne any code in __init___ that could affect the way how classifier/vectorizer is going to work,
+# Which means if you ever chagne any code in __init__ that could affect the way how classifier/vectorizer is going to work,
 # then you need to delete two files so that program recrete them. Those two files you need to delete are "classifier.joblib", and "vectorizer.joblib"
 # defined in constant.py.
 class ToxicityClassifier():

stopwords.txt

@@ -53,6 +53,7 @@ asked
 asking
 asks
 at
+atm
 av
 away
 b
@@ -61,6 +62,7 @@ backed
 backing
 backs
 be
+bc
 became
 because
 become
@@ -201,7 +203,9 @@ grouped
 grouping
 groups
 h
+ha
 had
+haha
 halves
 hardly
 has
@@ -237,6 +241,7 @@ however
 howevhowevhoif
 howsoever
 i
+ic
 ie
 if
 important
@@ -263,12 +268,17 @@ it
 its
 itself
 j
+jk
 just
 k
 keep
 keeps
 kg
 kind
+kk
+kkk
+kkkk
+kkkkk
 km
 knew
 know
@@ -290,9 +300,13 @@ lets
 like
 likely
 little
+lmao
+lmk
+lol
 long
 longer
 longest
+lul
 ltd
 m
 made
@@ -348,14 +362,14 @@ notwithstanding
 now
 nowadays
 nowhere
+np
 number
 numbers
 o
 of
 off
 often
 ok
-old
 older
 oldest
 on
@@ -482,6 +496,7 @@ stave
 staves
 still
 such
+sup
 supposing
 sure
 t
@@ -553,6 +568,7 @@ up
 upon
 upward
 upwards
+ur
 us
 use
 used
@@ -633,6 +649,7 @@ worst
 would
 wow
 x
+xdd
 y
 ye
 year
@@ -648,3 +665,4 @@ yourself
 yourselves
 yyyyyyyung
 z
+zzz