Spaces:
Sleeping
Sleeping
File size: 1,563 Bytes
b493a01 4709571 b493a01 4709571 b493a01 4709571 b493a01 4709571 b493a01 4709571 b493a01 4709571 b493a01 4709571 b493a01 4709571 b493a01 4709571 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 |
from collections import Counter
import streamlit as st
import srsly
def search(query):
results = []
for grant in grants:
if query in grant["tags"]:
results.append({"title": grant["title"], "tags": grant["tags"]})
st.session_state["results"] = results
st.header("Search 🔎 grants using MeSH 🔖")
st.sidebar.header("Information ℹ")
st.sidebar.write(
"A complete list of MeSH tags can be found here https://meshb.nlm.nih.gov/treeView"
)
st.sidebar.write("The grants data can be found https://www.threesixtygiving.org/")
st.sidebar.write(
"The model used to tag grants is https://huggingface.co/Wellcome/WellcomeBertMesh"
)
if "grants" not in st.session_state:
st.session_state["grants"] = list(srsly.read_jsonl("tagged_grants.jsonl"))
grants = st.session_state["grants"]
query = st.text_input("", value="Malaria")
st.button("Search 🔎", on_click=search, kwargs={"query": query})
if "results" in st.session_state:
st.caption("Related MeSH terms")
retrieved_tags = [tag for res in st.session_state["results"] for tag in res["tags"]]
most_common_tags = [tag for tag, _ in Counter(retrieved_tags).most_common(20)]
columns = st.columns(5)
for row_i in range(3):
for col_i, col in enumerate(columns):
with col:
tag_i = row_i * 5 + col_i
if tag_i < len(most_common_tags):
tag = most_common_tags[tag_i]
st.button(tag, on_click=search, kwargs={"query": tag})
st.table(st.session_state["results"])
|