|
import streamlit as st
|
|
from spacy import displacy
|
|
from Model.NER.VLSP2021.Predict_Ner import ViTagger
|
|
import re
|
|
|
|
import torch
|
|
|
|
|
|
|
|
|
|
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
|
|
|
def process_text(text):
|
|
|
|
processed_text = re.sub(r'\s+', ' ', text.strip())
|
|
return processed_text
|
|
|
|
def show_ner():
|
|
|
|
st.sidebar.title('Datasets')
|
|
dataset = st.sidebar.selectbox("Datasets", ("VLSP2016", "VLSP2021"))
|
|
st.header("NER")
|
|
text = st.text_area("Enter your text for NER:", height=300)
|
|
text = process_text(text)
|
|
if st.button("Process NER"):
|
|
if dataset == "VLSP2021":
|
|
tagger = ViTagger(model_path='E:/demo_datn/pythonProject1/Model/NER/VLSP2021/best_model.pt')
|
|
a = text
|
|
b = tagger.extract_entity_doc(a)
|
|
|
|
|
|
|
|
words_and_labels = b
|
|
|
|
words = [word for word, _ in words_and_labels]
|
|
|
|
entities = [{'start': sum(len(word) + 1 for word, _ in words_and_labels[:i]),
|
|
'end': sum(len(word) + 1 for word, _ in words_and_labels[:i + 1]), 'label': label} for
|
|
i, (word, label)
|
|
in enumerate(words_and_labels) if label != 'O']
|
|
|
|
|
|
|
|
html = displacy.render(
|
|
{"text": " ".join(words), "ents": entities, "title": None},
|
|
style="ent",
|
|
manual=True,
|
|
options={"colors": {"DATETIME-DATERANGE": "#66c2ff",
|
|
"LOCATION-GPE": "#ffcc99",
|
|
"O": None,
|
|
"QUANTITY-NUM": "#ffdf80",
|
|
"EVENT-CUL": "#bfbfbf",
|
|
"DATETIME": "#80ff80",
|
|
"PERSONTYPE": "#ff80ff",
|
|
"PERSON": "#bf80ff",
|
|
"QUANTITY-PER": "#80cccc",
|
|
"ORGANIZATION": "#ff6666",
|
|
"LOCATION-GEO": "#66cc66",
|
|
"LOCATION-STRUC": "#cccc66",
|
|
"PRODUCT-COM": "#ffff66",
|
|
"DATETIME-DATE": "#66cccc",
|
|
"QUANTITY-DIM": "#6666ff",
|
|
"PRODUCT": "#cc6666",
|
|
"QUANTITY": "#6666cc",
|
|
"DATETIME-DURATION": "#9966ff",
|
|
"QUANTITY-CUR": "#ff9966",
|
|
"DATETIME-TIME": "#cdbf93",
|
|
"QUANTITY-TEM": "#cc9966",
|
|
"DATETIME-TIMERANGE": "#cc8566",
|
|
"EVENT-GAMESHOW": "#8c8c5a",
|
|
"QUANTITY-AGE": "#70db70",
|
|
"QUANTITY-ORD": "#e699ff",
|
|
"PRODUCT-LEGAL": "#806699",
|
|
"LOCATION": "#993366",
|
|
"ORGANIZATION-MED": "#339933",
|
|
"URL": "#ff4d4d",
|
|
"PHONENUMBER": "#99cc99",
|
|
"ORGANIZATION-SPORTS": "#6666ff",
|
|
"EVENT-SPORT": "#ffff80",
|
|
"SKILL": "#b38f66",
|
|
"EVENT-NATURAL": "#ff9966",
|
|
"ADDRESS": "#cc9966",
|
|
"IP": "#b38f66",
|
|
"EMAIL": "#cc8566",
|
|
"ORGANIZATION-STOCK": "#666633",
|
|
"DATETIME-SET": "#70db70",
|
|
"PRODUCT-AWARD": "#e699ff",
|
|
"MISCELLANEOUS": "#806699",
|
|
"LOCATION-GPE-GEO": "#99ffff"}}
|
|
)
|
|
|
|
st.markdown(html, unsafe_allow_html=True)
|
|
elif dataset == "VLSP2016":
|
|
tagger = ViTagger(model_path='E:/demo_datn/pythonProject1/Model/NER/VLSP2016/best_model.pt')
|
|
a = text
|
|
b = tagger.extract_entity_doc(a)
|
|
|
|
|
|
|
|
words_and_labels = b
|
|
|
|
words = [word for word, _ in words_and_labels]
|
|
|
|
entities = [{'start': sum(len(word) + 1 for word, _ in words_and_labels[:i]),
|
|
'end': sum(len(word) + 1 for word, _ in words_and_labels[:i + 1]), 'label': label} for
|
|
i, (word, label)
|
|
in enumerate(words_and_labels) if label != 'O']
|
|
|
|
|
|
|
|
html = displacy.render(
|
|
{"text": " ".join(words), "ents": entities, "title": None},
|
|
style="ent",
|
|
manual=True,
|
|
options={"colors": {"MISC": "#806699",
|
|
"ORG": "#ff6666",
|
|
"LOC": "#66cc66",
|
|
"PER": "#bf80ff",
|
|
"O": None}}
|
|
)
|
|
|
|
st.markdown(html, unsafe_allow_html=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|