File size: 1,102 Bytes
673bdce
 
 
 
 
 
93f465d
 
b48f7ea
 
fc36661
054a10b
b48f7ea
a01266f
ee93acc
31a82df
ee93acc
 
 
 
054a10b
31a82df
6415a43
054a10b
fc36661
a01266f
31a82df
 
a01266f
054a10b
fc36661
054a10b
a01266f
673bdce
 
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
import streamlit as st

def show_model():
  return
  
def main():
  st.title("Semantic Search for Datasets Using Sentence Transformers")
  st.write("A case study for the National Centers for Environmental Information (NCEI)")
  st.image("noaa_logo.png", width=150)
  
  st.write("## Goal: search for datasets in NCEI's Archive using natural language queries")
  st.write("[Repo](https://github.com/myrandaGoesToSpace/semantic-search-datasets)")
  
  st.image("pres-whatisnoaa.png")
  
  st.write("### Uses service called OneStop for data search")
  st.write("**Problems:**")
  st.write("- Uses keyword search -- not robust to natural language queries")
  #st.image("pres-onestop.png")
  #st.image("pres-problems.png")
  
  st.write("## The Model: [Sentence Transformers](https://huggingface.co/sentence-transformers/multi-qa-MiniLM-L6-cos-v1)")
  st.image("pres-sentencetransformers.png")
  
  st.write("## Project Data")
  st.image("pres-metadata.png")
  
  st.write("## The Process")
  st.image("pres-creatingse.png")
  
  st.write("## Results and Demo")
  
  st.image("pres-futureplans.png")

main()