Spaces:
Running
Running
| import spacy | |
| import json | |
| from datetime import datetime | |
| import streamlit as st | |
| 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.title("Annotation Tool") | |
| # Add Materialize CSS | |
| st.markdown('<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/materialize/1.0.0/css/materialize.min.css">', unsafe_allow_html=True) | |
| 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:") | |
| for token in annotated_text: | |
| st.write(f"Token: {token['token']}") | |
| st.write(f"POS: {token['pos']}") | |
| st.write(f"Lemma: {token['lemma']}") | |
| st.write(f"Aspect: {', '.join(token['aspect'])}") | |
| st.write(f"Tense: {', '.join(token['tense'])}") | |
| st.write(f"VerbForm: {', '.join(token['verbForm'])}") | |
| st.write(f"Voice: {', '.join(token['voice'])}") | |
| st.write(f"Mood: {', '.join(token['mood'])}") | |
| st.write(f"Number: {', '.join(token['number'])}") | |
| st.write(f"Person: {', '.join(token['person'])}") | |
| st.write(f"Case: {', '.join(token['case'])}") | |
| st.write(f"Gender: {', '.join(token['gender'])}") | |
| st.write("-----") | |
| else: | |
| st.warning("Please enter some text.") | |
| if st.button("Save Modifications as JSON"): | |
| if annotated_text: | |
| # Generate a unique file name using a timestamp | |
| timestamp = datetime.now().strftime("%Y%m%d%H%M%S") | |
| filename = f'annotations_{timestamp}.json' | |
| save_annotations_as_json(annotated_text, filename) | |
| st.success(f"Annotations saved as {filename}") | |
| # Add Materialize JavaScript | |
| st.markdown('<script src="https://cdnjs.cloudflare.com/ajax/libs/materialize/1.0.0/js/materialize.min.js"></script>', unsafe_allow_html=True) |