VulgarLatin / app.py
bestroi's picture
Update app.py
3b5f32b
import spacy
import json
from datetime import datetime
import streamlit as st
import pandas as pd
nlp = spacy.load('la_core_web_lg')
def tokenize_with_spacy(text):
doc = nlp(text)
return [token.text for token in doc]
def annotate_text(segmented_text):
annotated_tokens = []
for token in segmented_text:
doc = nlp(token)
annotated_token = {
'token': token,
'pos': str(doc[0].pos_),
'lemma': str(doc[0].lemma_),
'aspect': ', '.join(doc[0].morph.get("Aspect", default=[""])),
'tense': ', '.join(doc[0].morph.get("Tense", default=[""])),
'verbForm': ', '.join(doc[0].morph.get("VerbForm", default=[""])),
'voice': ', '.join(doc[0].morph.get("Voice", default=[""])),
'mood': ', '.join(doc[0].morph.get("Mood", default=[""])),
'number': ', '.join(doc[0].morph.get("Number", default=[""])),
'person': ', '.join(doc[0].morph.get("Person", default=[""])),
'case': ', '.join(doc[0].morph.get("Case", default=[""])),
'gender': ', '.join(doc[0].morph.get("Gender", default=[""]))
}
annotated_tokens.append(annotated_token)
return annotated_tokens
def save_annotations_as_json(annotated_text, filename):
with open(filename, 'w', encoding='utf-8') as json_file:
json.dump(annotated_text, json_file, ensure_ascii=False, indent=4)
st.set_page_config(layout="wide")
st.title("Annotation Tool")
text = st.text_area("Text")
if st.button("Annotate"):
if text:
segmented_text = tokenize_with_spacy(text)
annotated_text = annotate_text(segmented_text)
st.subheader("Segmented Text:")
st.write(segmented_text)
st.subheader("Annotated Text:")
# Create a DataFrame from the annotated text
df = pd.DataFrame(annotated_text)
st.write(df)
if st.button("Save Modifications as JSON"):
save_annotations_as_json(df.to_dict(orient='records'), 'annotations.json')
st.success("Annotations saved as annotations.json")
else:
st.warning("Please enter some text.")