import spacy
import streamlit as st
from spacy_streamlit import visualize_ner

try:
    spacy.load("en_core_web_sm")
except:
    spacy.cli.download("en_core_web_sm")

st.write("""
# Named Entity Recognition for Technology and Science Keywords.
""")


nlp = spacy.load("en_core_web_sm")
ruler = nlp.add_pipe("entity_ruler", before="ner")
ruler.from_disk("tokens.json")

description = "Engineering Type	Median Salary (May 2020)	Job Growth Rate (2020-30)
Petroleum Engineers	$137,330	8%
Computer Hardware Engineers	$119,560	2%
Aerospace Engineers	$118,610	8%
Nuclear Engineers	$116,140	-8%
Chemical Engineers	$108,540	9%
Electrical and Electronics Engineers	$103,390	7%
Materials Engineers	$95,640	8%
Marine Engineers	$95,440	4%
Health and Safety Engineers	$94,240	6%
Mining and Geological Engineers	$93,800	4%
Biomedical Engineers	$92,620	6%
Environmental Engineers	$92,120	4%
Mechanical Engineers	$90,160	7%
Industrial Engineers	$88,950	14%
Civil Engineers	$88,570	8%
Agricultural Engineers	$84,410	5%"

text = st.text_area(label='Engineering Description', value=description, placeholder='Enter Text for Technology or Engineering')

doc = nlp(text)
visualize_ner(doc, labels=nlp.get_pipe("ner").labels)

st.subheader('Keywords')
st.json(
    [{'label': entity.label_, 'text': entity.text, 'start': entity.start, 'end': entity.end} \
        for entity in doc.ents if entity.ent_id_ == 'SKILLS']
)