Demo.py

import streamlit as st
import sparknlp
import os
import pandas as pd

from sparknlp.base import *
from sparknlp.annotator import *
from pyspark.ml import Pipeline
from sparknlp.pretrained import PretrainedPipeline

# Page configuration
st.set_page_config(
    layout="wide", 
    page_title="Spark NLP Demos App", 
    initial_sidebar_state="auto"
)

# CSS for styling
st.markdown("""
    <style>
        .main-title {
            font-size: 36px;
            color: #4A90E2;
            font-weight: bold;
            text-align: center;
        }
        .section p, .section ul {
            color: #666666;
        }
    </style>
""", unsafe_allow_html=True)

@st.cache_resource
def init_spark():
    return sparknlp.start()

@st.cache_resource
def create_pipeline(model):
  documentAssembler = DocumentAssembler()\
  .setInputCol("text")\
  .setOutputCol("document")
    
  use = UniversalSentenceEncoder.pretrained()\
  .setInputCols(["document"])\
  .setOutputCol("sentence_embeddings")

  sentimentdl = ClassifierDLModel.pretrained(model)\
    .setInputCols(["sentence_embeddings"])\
    .setOutputCol("sentiment")

  nlpPipeline = Pipeline(stages = [documentAssembler, use, sentimentdl])
  
  return nlpPipeline

def fit_data(pipeline, data):
    empty_df = spark.createDataFrame([['']]).toDF('text')
    pipeline_model = pipeline.fit(empty_df)
    model = LightPipeline(pipeline_model)
    results = model.fullAnnotate(data)[0]

    return results['sentiment'][0].result

# Set up the page layout
st.markdown('<div class="main-title">Detect Sarcastic Tweets with Spark NLP</div>', unsafe_allow_html=True)

# Sidebar content
model = st.sidebar.selectbox(
    "Choose the pretrained model",
    ["classifierdl_use_sarcasm"],
    help="For more info about the models visit: https://sparknlp.org/models"
)

# Reference notebook link in sidebar
link = """
<a href="https://colab.research.google.com/github/JohnSnowLabs/spark-nlp-workshop/blob/master/tutorials/streamlit_notebooks/SENTIMENT_EN_SARCASM.ipynb">
    <img src="https://colab.research.google.com/assets/colab-badge.svg" style="zoom: 1.3" alt="Open In Colab"/>
</a>
"""
st.sidebar.markdown('Reference notebook:')
st.sidebar.markdown(link, unsafe_allow_html=True)

# Load examples
examples = [
  "Love getting home from work knowing that in less than 8hours you're getting up to go back there again.",
  "Oh my gosh! Can you imagine @JessieJ playing piano on her tour while singing a song. I would die and go to heaven. #sheisanangel",
  "Dear Teva, thank you for waking me up every few hours by howling. Your just trying to be mother natures alarm clock.",
  "The United States is a signatory to this international convention",
  "If I could put into words how much I love waking up at am on Tuesdays I would",
  "@pdomo Don't forget that Nick Foles is also the new Tom Brady. What a preseason! #toomanystudQBs #thankgodwedonthavetebow",
  "I cant even describe how excited I am to go cook noodles for hours",
  "@Will_Piper should move back up fella. I'm already here... On my own... Having loads of fun",
  "Tweeting at work... Having sooooo much fun and honestly not bored at all #countdowntillfinish",
  "I can do what I want to. I play by my own rules"
]

selected_text = st.selectbox("Select a sample", examples)
custom_input = st.text_input("Try it for yourself!")

if custom_input:
    selected_text = custom_input
elif selected_text:
    selected_text = selected_text

st.subheader('Selected Text')
st.write(selected_text)

st.subheader('Selected Text')
st.write(selected_text)

# Initialize Spark and create pipeline
spark = init_spark()
pipeline = create_pipeline(model)
output = fit_data(pipeline, selected_text)

# Display output sentence
if output in ['neutral', 'normal']:
  st.markdown("""<h3>This seems like <span style="color: #209DDC">{}</span> news. <span style="font-size:35px;">&#128578;</span></h3>""".format(output), unsafe_allow_html=True)
elif output == 'sarcasm':
  st.markdown("""<h3>This seems like a <span style="color: #B64434">{}</span> tweet. <span style="font-size:35px;">&#128579;</span></h3>""".format('sarcastic'), unsafe_allow_html=True)