updated autocomplete for nearest neighbours
Browse files- app.py +6 -5
- autocomplete.py +37 -0
- corpora/compass_filtered.pkl.gz +3 -0
app.py
CHANGED
@@ -2,6 +2,7 @@ import streamlit as st
|
|
2 |
from streamlit_option_menu import option_menu
|
3 |
from word2vec import *
|
4 |
import pandas as pd
|
|
|
5 |
|
6 |
st.set_page_config(page_title="Ancient Greek Word2Vec", layout="centered")
|
7 |
|
@@ -13,14 +14,14 @@ active_tab = option_menu(None, ["Nearest neighbours", "Cosine similarity", "3D g
|
|
13 |
if active_tab == "Nearest neighbours":
|
14 |
st.write("### TO DO: add description of function")
|
15 |
col1, col2 = st.columns(2)
|
16 |
-
|
17 |
-
|
18 |
-
|
|
|
19 |
|
20 |
with st.container():
|
21 |
with col1:
|
22 |
-
|
23 |
-
word = st.multiselect("Enter a word", all_words, default=["πατήρ"])
|
24 |
|
25 |
with col2:
|
26 |
time_slice = st.selectbox("Time slice", ["Archaic", "Classical", "Hellenistic", "Early Roman", "Late Roman"])
|
|
|
2 |
from streamlit_option_menu import option_menu
|
3 |
from word2vec import *
|
4 |
import pandas as pd
|
5 |
+
from autocomplete import *
|
6 |
|
7 |
st.set_page_config(page_title="Ancient Greek Word2Vec", layout="centered")
|
8 |
|
|
|
14 |
if active_tab == "Nearest neighbours":
|
15 |
st.write("### TO DO: add description of function")
|
16 |
col1, col2 = st.columns(2)
|
17 |
+
|
18 |
+
# Load the compressed word list
|
19 |
+
compressed_word_list_filename = 'corpora/compass_filtered.pkl.gz'
|
20 |
+
all_words = load_compressed_word_list(compressed_word_list_filename)
|
21 |
|
22 |
with st.container():
|
23 |
with col1:
|
24 |
+
word = st.multiselect("Enter a word", all_words)
|
|
|
25 |
|
26 |
with col2:
|
27 |
time_slice = st.selectbox("Time slice", ["Archaic", "Classical", "Hellenistic", "Early Roman", "Late Roman"])
|
autocomplete.py
ADDED
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import pickle
|
2 |
+
import gzip
|
3 |
+
|
4 |
+
|
5 |
+
def get_unique_words(corpus_filename):
|
6 |
+
"""
|
7 |
+
Get a list of unique words from a corpus file
|
8 |
+
"""
|
9 |
+
unique_words = set()
|
10 |
+
with open(corpus_filename, 'r', encoding='utf-8') as file:
|
11 |
+
for line in file:
|
12 |
+
words = line.strip().split()
|
13 |
+
unique_words.update(words)
|
14 |
+
return list(unique_words)
|
15 |
+
|
16 |
+
|
17 |
+
def save_compressed_word_list(words, filename):
|
18 |
+
"""
|
19 |
+
Save a list of words to a compressed file
|
20 |
+
"""
|
21 |
+
with gzip.open(filename, 'wb') as file:
|
22 |
+
pickle.dump(words, file)
|
23 |
+
|
24 |
+
|
25 |
+
def load_compressed_word_list(filename):
|
26 |
+
"""
|
27 |
+
Load a list of words from a compressed file
|
28 |
+
"""
|
29 |
+
with gzip.open(filename, 'rb') as file:
|
30 |
+
return pickle.load(file)
|
31 |
+
|
32 |
+
|
33 |
+
def get_autocomplete(input_word=" ", all_words=" "):
|
34 |
+
"""
|
35 |
+
Get a list of words that start with the input word
|
36 |
+
"""
|
37 |
+
return [word for word in all_words if word.startswith(input_word)]
|
corpora/compass_filtered.pkl.gz
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a02f7e61330045219098cda092f3edb23650decc70d60db6be2de267e8c7a925
|
3 |
+
size 163769
|