GreatestParrot
commited on
Commit
•
3789923
1
Parent(s):
8543f72
add model
Browse files- .gitattributes +1 -0
- app.py +37 -0
- flagged/log.csv +2 -0
- proposals_clean.xlsx +3 -0
- proposals_emb.xlsx +3 -0
- requirements.txt +5 -0
.gitattributes
CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
*.xlsx filter=lfs diff=lfs merge=lfs -text
|
app.py
ADDED
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from sentence_transformers import SentenceTransformer, util
|
2 |
+
import numpy as np
|
3 |
+
import pandas as pd
|
4 |
+
import gradio as gr
|
5 |
+
|
6 |
+
#Load the model
|
7 |
+
model = SentenceTransformer('sentence-transformers/msmarco-bert-base-dot-v5')
|
8 |
+
doc_emb = pd.read_excel("proposals_emb.xlsx", usecols=lambda x: str(x).isnumeric())
|
9 |
+
df = pd.read_excel("proposals_clean.xlsx")
|
10 |
+
|
11 |
+
|
12 |
+
def cosine(u, v):
|
13 |
+
res = np.dot(u, v) / (np.linalg.norm(u) * np.linalg.norm(v))
|
14 |
+
return res
|
15 |
+
|
16 |
+
|
17 |
+
def form_link(post_id):
|
18 |
+
return f"https://kusama.polkassembly.io/referenda/{post_id}"
|
19 |
+
|
20 |
+
|
21 |
+
def processing(query):
|
22 |
+
query_emb = model.encode(query)
|
23 |
+
print(doc_emb.columns)
|
24 |
+
doc_emb['sim1'] = doc_emb.apply(lambda row: cosine(row, query_emb), axis=1)
|
25 |
+
sim = doc_emb.nlargest(5, 'sim1').index
|
26 |
+
res = df.iloc[sim][['content', 'status']]
|
27 |
+
doc_emb.drop(columns=["sim1"], inplace=True)
|
28 |
+
|
29 |
+
out = [f"[Proposal #{row[0]}]({form_link(row[0])}) - {row[1]['status']}: {ind+1}" for ind, row in enumerate(res.iterrows())]
|
30 |
+
markdown_string = ""
|
31 |
+
for i in range(len(out)):
|
32 |
+
markdown_string += f"{i+1}. {out[i]}\n"
|
33 |
+
return markdown_string
|
34 |
+
|
35 |
+
|
36 |
+
iface = gr.Interface(processing, "text", "markdown")
|
37 |
+
iface.launch()
|
flagged/log.csv
ADDED
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
1 |
+
name,output,flag,username,timestamp
|
2 |
+
sdafasdfasd,Hello sdafasdfasd!!,,,2023-11-07 14:09:04.894703
|
proposals_clean.xlsx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:87e1dfc2c1655932b4970bc266770aac58c5f07956b96c8699b0198509847c3c
|
3 |
+
size 270107
|
proposals_emb.xlsx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f2c3482a474475ca1e36a538c7b5957d66b6a10c7e801cf74616071750190dd4
|
3 |
+
size 3020034
|
requirements.txt
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
sentence-transformers
|
2 |
+
pandas
|
3 |
+
numpy
|
4 |
+
gradio
|
5 |
+
openpyxl
|