def test_lemmatized_text():
"""Test lemmatized text works"""
text = "She loves dogs"
output = az.lemmatized_text(text)
expect = "love dog"
print(output)
assert output == expect
def interactive():
"""Page to allow nlp analysis from user input"""
input_text = st.text_area("Enter text", "Type here")
token_cb = st.checkbox("Show tokens")
ner_cb = st.checkbox("Show named entities")
sentiment_cb = st.checkbox("Show sentiment")
summary_cb = st.checkbox("Show Summary")
# st.success("Running Analysis")
# if st.button("Analysis"):
if token_cb:
tokens = az.tokenize(input_text)
st.write(tokens)
if ner_cb:
doc = az.get_nlp(input_text)
named_entities = az.named_entity_recognization(input_text)
if len(named_entities) > 0:
html = spacy.displacy.render(doc, style="ent")
# Newlines seem to mess with the rendering
html = html.replace("\n", " ")
HTML_WRAPPER = """<div style="overflow-x: auto; border: 1px solid \
#e6e9ef; border-radius: 0.25rem; padding: 1rem; margin-bottom: 2.5rem">\
{}</div>"""
st.write(HTML_WRAPPER.format(html), unsafe_allow_html=True)
else:
st.info("No named entity recognized")
if sentiment_cb:
sentiments = TextBlob(az.lemmatized_text(input_text))
st.write(sentiments.sentiment)
if summary_cb:
summaries = sz.summarize_text(input_text)
st.write(summaries)
def sentiment():
"""Main function for sentiment analysis."""
senti_df = main_df.copy(deep=True)
# Initializing the new columns with a numpy array, so the entire series is returned
senti_df[cts.POSITIVE], senti_df[cts.NEGATIVE] = az.top_polarized_word(
senti_df[cts.TOKEN].values)
# calculate overall sentiment from the combined text
senti_df[cts.SENTI] = senti_df[cts.COMBINED].apply(
lambda x: TextBlob(az.lemmatized_text(x)).sentiment.polarity)
senti_df = ut.return_assignment(senti_df, assign_id, assignments)
senti_type = st.sidebar.selectbox("Type of sentiment analysis",
["Overall", "Student", "Question"])
if not assignments:
st.warning("Please select an assignment for the analysis")
elif senti_type == "Overall":
st.sidebar.success(
'To continue see individual sentiment analysis select "Student"')
st.header(f"Overall sentiment polarity in **{assignment_string}**")
overall_senti(senti_df)
elif senti_type == "Student":
st.header(
f"View sentiment by individual students in **{assignment_string}**"
)
student_senti(senti_df)
elif senti_type == "Question":
st.header(
f"View sentiment by individual questions in **{assignment_string}**"
)
question_senti(senti_df)
def interactive():
"""Page to allow nlp analysis from user input."""
input_text = st.text_area("Enter text", "Type here")
token_cb = st.checkbox("Show tokens")
ner_cb = st.checkbox("Show named entities")
sentiment_cb = st.checkbox("Show sentiment")
summary_cb = st.checkbox("Show Summary")
# st.success("Running Analysis")
# if st.button("Analysis"):
if token_cb:
tokens = az.tokenize(input_text)
st.write(tokens)
if ner_cb:
displacy_renderer(az.get_nlp(input_text))
if sentiment_cb:
sentiments = TextBlob(az.lemmatized_text(input_text))
st.write(sentiments.sentiment)
if summary_cb:
summaries = sz.summarize_text(input_text)
st.write(summaries)
def sentiment():
"""main function for sentiment analysis"""
senti_df = main_df.copy(deep=True)
# calculate overall sentiment from the combined text
senti_df[cts.SENTI] = senti_df["combined"].apply(
lambda x: TextBlob(az.lemmatized_text(x)).sentiment.polarity)
senti_df = senti_df[senti_df[assign_id].isin(assignments)]
senti_type = st.sidebar.selectbox("Type of sentiment analysis",
["Overall", "Student", "Question"])
if senti_type == "Overall":
st.sidebar.success(
'To continue see individual sentiment analysis select "Student"')
st.header(f"Overall sentiment polarity in **{assign_text}**")
overall_senti(senti_df)
elif senti_type == "Student":
st.header(
f"View sentiment by individual students in **{assign_text}**")
student_senti(senti_df)
elif senti_type == "Question":
st.header(
f"View sentiment by individual questions in **{assign_text}**")
question_senti(senti_df)