import streamlit as st from transformers import pipeline st.set_page_config(page_title="Common NLP Tasks") st.title("Common NLP Tasks") st.subheader("Use the menu on the left to select a NLP task to do (click on > if closed).") expander = st.sidebar.expander("About") expander.write("This web app allows you to perform common Natural Language Processing tasks, select a task below to get started.") st.sidebar.header("What will you like to do?") option = st.sidebar.radio("", ["Extractive question answering", "Text summarization", "Text generation", "Sentiment analysis"]) @st.cache(show_spinner=False, allow_output_mutation=True) def question_model(): model_name = "deepset/roberta-base-squad2" question_answerer = pipeline(model=model_name, tokenizer=model_name, revision="v1.0", task="question-answering") return question_answerer @st.cache(show_spinner=False, allow_output_mutation=True) def summarization_model(): summarizer = pipeline("summarization") return summarizer @st.cache(show_spinner=False, allow_output_mutation=True) def generation_model(): generator = pipeline("text-generation") return generator @st.cache(show_spinner=False, allow_output_mutation=True) def sentiment_model(): sentiment_analysis = pipeline("sentiment-analysis") return sentiment_analysis if option == "Extractive question answering": st.markdown("

Extract answer from text

", unsafe_allow_html=True) source = st.radio("How would you like to start? Choose an option below", ["I want to input some text", "I want to upload a file"]) if source == "I want to input some text": with open("sample.txt", "r") as text_file: sample_text = text_file.read() context = st.text_area('Use the example below or input your own text in English (10,000 characters max)', value=sample_text, max_chars=10000, height=330) question = st.text_input(label='Enter your question') button = st.button('Get answer') if button: question_answerer = question_model() with st.spinner(text="Getting answer..."): answer = question_answerer(context=context, question=question) answer = answer["answer"] st.write(f"Answer ---> {answer}") elif source == "I want to upload a file": uploaded_file = st.file_uploader("Choose a .txt file to upload", type=["txt"]) if uploaded_file is not None: context = st.text_area("", value=uploaded_file.read(), height=330) question = st.text_input(label="Enter your question") button = st.button("Get answer") if button: question_answerer = question_model() with st.spinner(text="Getting answer..."): answer = question_answerer(context=context, question=question) st.write(f"Answer ---> {answer}") elif option == 'Text summarization': st.markdown("

Summarize text

", unsafe_allow_html=True) sample_text = "sample text" source = st.radio("How would you like to start? Choose an option below", ["I want to input some text", "I want to upload a file"]) if source == "I want to input some text": text = st.text_area('Input a text in English (between 1,000 and 10,000 characters)', value=sample_text, max_chars=10000, height=330) button = st.button('Get summary') if button: summarizer = summarization_model() with st.spinner(text="Summarizing text..."): summary = summarizer(text, max_length=130, min_length=30) st.write(summary) elif source == "I want to upload a file": uploaded_file = st.file_uploader("Choose a .txt file to upload", type=["txt"]) button = st.button('Get summary') if button: summarizer = summarization_model() with st.spinner(text="Summarizing text..."): summary = summarizer(text, max_length=130, min_length=30) st.write(summary) elif option == 'Text generation': st.markdown("

Generate text

", unsafe_allow_html=True) text = st.text_input(label='Enter one line of text and let the NLP model generate the rest for you') button = st.button('Generate text') if button: generator = generation_model() with st.spinner(text="Generating text..."): generated_text = generator(text, max_length=50) st.write(generated_text[0]["generated_text"]) elif option == 'Sentiment analysis': st.markdown("

Classify review

", unsafe_allow_html=True) text = st.text_input(label='Enter a sentence to get its sentiment analysis') button = st.button('Get sentiment analysis') if button: sentiment_analysis = sentiment_model() with st.spinner(text="Getting sentiment analysis..."): sentiment = sentiment_analysis(text) st.write(sentiment[0]["label"])