put nearest neighbours function into a form (input process is faster now)
Browse files
app.py
CHANGED
@@ -24,26 +24,19 @@ lemma_dict = json.load(open('lsj_dict.json', 'r'))
|
|
24 |
|
25 |
# Nearest neighbours tab
|
26 |
if active_tab == "Nearest neighbours":
|
27 |
-
st.write("### TO DO: add description of function")
|
28 |
-
col1, col2 = st.columns(2)
|
29 |
|
30 |
# Load the compressed word list
|
31 |
compressed_word_list_filename = 'corpora/compass_filtered.pkl.gz'
|
32 |
all_words = load_compressed_word_list(compressed_word_list_filename)
|
33 |
eligible_models = ["Archaic", "Classical", "Hellenistic", "Early Roman", "Late Roman"]
|
34 |
|
35 |
-
|
36 |
-
st.
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
word = st.multiselect("Enter a word", all_words, max_selections=1)
|
41 |
-
if len(word) > 0:
|
42 |
-
word = word[0]
|
43 |
-
|
44 |
-
# Check which models contain the word
|
45 |
-
eligible_models = check_word_in_models(word)
|
46 |
|
|
|
47 |
|
48 |
models = st.multiselect(
|
49 |
"Select models to search for neighbours",
|
@@ -51,49 +44,45 @@ if active_tab == "Nearest neighbours":
|
|
51 |
)
|
52 |
n = st.slider("Number of neighbours", 1, 50, 15)
|
53 |
|
54 |
-
nearest_neighbours_button = st.
|
55 |
|
56 |
-
|
57 |
-
if
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
58 |
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
nearest_neighbours = get_nearest_neighbours(word, n, models)
|
67 |
|
68 |
-
|
69 |
-
|
70 |
-
|
71 |
-
|
72 |
-
|
73 |
-
|
74 |
-
nearest_neighbours[model],
|
75 |
-
columns = ['Word', 'Cosine Similarity']
|
76 |
)
|
77 |
-
|
78 |
-
all_dfs.append((model, df))
|
79 |
-
st.table(df)
|
80 |
-
|
81 |
-
|
82 |
-
# Store content in a temporary file
|
83 |
-
tmp_file = store_df_in_temp_file(all_dfs)
|
84 |
-
|
85 |
-
# Open the temporary file and read its content
|
86 |
-
with open(tmp_file, "rb") as file:
|
87 |
-
file_byte = file.read()
|
88 |
-
|
89 |
-
# Create download button
|
90 |
-
st.download_button(
|
91 |
-
"Download results",
|
92 |
-
data=file_byte,
|
93 |
-
file_name = f'nearest_neighbours_{word}_TEST.xlsx',
|
94 |
-
mime='application/octet-stream'
|
95 |
-
)
|
96 |
-
|
97 |
|
98 |
|
99 |
# Cosine similarity tab
|
|
|
24 |
|
25 |
# Nearest neighbours tab
|
26 |
if active_tab == "Nearest neighbours":
|
|
|
|
|
27 |
|
28 |
# Load the compressed word list
|
29 |
compressed_word_list_filename = 'corpora/compass_filtered.pkl.gz'
|
30 |
all_words = load_compressed_word_list(compressed_word_list_filename)
|
31 |
eligible_models = ["Archaic", "Classical", "Hellenistic", "Early Roman", "Late Roman"]
|
32 |
|
33 |
+
with st.form("nn_form"):
|
34 |
+
st.markdown("## Nearest Neighbours")
|
35 |
+
target_word = st.multiselect("Enter a word", all_words, max_selections=1)
|
36 |
+
if len(target_word) > 0:
|
37 |
+
target_word = target_word[0]
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
|
39 |
+
eligible_models = check_word_in_models(target_word)
|
40 |
|
41 |
models = st.multiselect(
|
42 |
"Select models to search for neighbours",
|
|
|
44 |
)
|
45 |
n = st.slider("Number of neighbours", 1, 50, 15)
|
46 |
|
47 |
+
nearest_neighbours_button = st.form_submit_button("Find nearest neighbours", on_click = click_nn_button)
|
48 |
|
49 |
+
if nearest_neighbours_button:
|
50 |
+
if validate_nearest_neighbours(target_word, n, models) == False:
|
51 |
+
st.error('Please fill in all fields')
|
52 |
+
else:
|
53 |
+
# Rewrite models to list of all loaded models
|
54 |
+
models = load_selected_models(models)
|
55 |
+
|
56 |
+
nearest_neighbours = get_nearest_neighbours(target_word, n, models)
|
57 |
+
|
58 |
+
all_dfs = []
|
59 |
+
|
60 |
+
# Create dataframes
|
61 |
+
for model in nearest_neighbours.keys():
|
62 |
+
st.write(f"### {model}")
|
63 |
+
df = pd.DataFrame(
|
64 |
+
nearest_neighbours[model],
|
65 |
+
columns = ['Word', 'Cosine Similarity']
|
66 |
+
)
|
67 |
+
|
68 |
+
all_dfs.append((model, df))
|
69 |
+
st.table(df)
|
70 |
|
71 |
+
|
72 |
+
# Store content in a temporary file
|
73 |
+
tmp_file = store_df_in_temp_file(all_dfs)
|
74 |
+
|
75 |
+
# Open the temporary file and read its content
|
76 |
+
with open(tmp_file, "rb") as file:
|
77 |
+
file_byte = file.read()
|
|
|
78 |
|
79 |
+
# Create download button
|
80 |
+
st.download_button(
|
81 |
+
"Download results",
|
82 |
+
data=file_byte,
|
83 |
+
file_name = f'nearest_neighbours_{target_word}_TEST.xlsx',
|
84 |
+
mime='application/octet-stream'
|
|
|
|
|
85 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
86 |
|
87 |
|
88 |
# Cosine similarity tab
|