app.py

# # import streamlit as st
# # import transformers
# # import torch

# # # Load the model and tokenizer
# # model = transformers.AutoModelForSequenceClassification.from_pretrained("DeeeTeeee01/twitter-xlm-roberta-base-sentiment_dee")
# # tokenizer = transformers.AutoTokenizer.from_pretrained("DeeeTeeee01/twitter-xlm-roberta-base-sentiment_dee")

# # # Define the function for sentiment analysis
# # @st.cache_resource
# # def predict_sentiment(text):
# #     # Load the pipeline.
# #     pipeline = transformers.pipeline("sentiment-analysis")

# #     # Predict the sentiment.
# #     prediction = pipeline(text)
# #     sentiment = prediction[0]["label"]
# #     score = prediction[0]["score"]

# #     return sentiment, score

# # # Setting the page configurations
# # st.set_page_config(
# #     page_title="Sentiment Analysis App",
# #     page_icon=":smile:",
# #     layout="wide",
# #     initial_sidebar_state="auto",
# # )

# # # Add description and title
# # st.write("""
# # # Predict if your text is  Positive, Negative or Nuetral ...
# # Please type your text and press ENTER key to know if your text is positive, negative, or neutral sentiment!
# # """)


# # # Add image
# # image = st.image("sentiment.jpeg", width=400)

# # # Get user input
# # text = st.text_input("Type here:")

# # # Define the CSS style for the app
# # st.markdown(
# # """
# # <style>
# # body {
# #     background-color: #f5f5f5;
# # }
# # h1 {
# #     color: #4e79a7;
# # }
# # </style>
# # """,
# # unsafe_allow_html=True
# # )

# # # Show sentiment output
# # if text:
# #     sentiment, score = predict_sentiment(text)
# #     if sentiment == "Positive":
# #         st.success(f"The sentiment is {sentiment} with a score of {score*100:.2f}%!")
# #     elif sentiment == "Negative":
# #         st.error(f"The sentiment is {sentiment} with a score of {score*100:.2f}%!")
# #     else:
# #         st.warning(f"The sentiment is {sentiment} with a score of {score*100:.2f}%!")

# import streamlit as st
# import transformers
# import torch

# # Load the model and tokenizer
# model = transformers.AutoModelForSequenceClassification.from_pretrained("DeeeTeeee01/mytest_trainer_roberta-base")
# tokenizer = transformers.AutoTokenizer.from_pretrained("DeeeTeeee01/mytest_trainer_roberta-base")

# # Define the function for sentiment analysis
# @st.cache_resource
# def predict_sentiment(text):
#     # Load the pipeline
#     pipeline = transformers.pipeline("sentiment-analysis")

    
#     # Predict the sentiment
#     prediction = pipeline(text)
#     sentiment = prediction[0]["label"]
#     score = prediction[0]["score"]

#     return sentiment, score

# # Setting the page configurations
# st.set_page_config(
#     page_title="Sentiment Analysis App",
#     page_icon=":smile:",
#     layout="wide",
#     initial_sidebar_state="auto",
# )

# # Add description and title
# st.write("""
# # Predict if your text is Positive, Negative or Neutral ...
# Please type your text and click the Predict button to know if your text has a positive, negative or neutral sentiment!
# """)

# # Add image
# image = st.image("sentiment.jpeg", width=400)

# # Get user input
# text = st.text_input("Type here:")

# # Add Predict button
# predict_button = st.button("Predict")

# # Define the CSS style for the app
# st.markdown(
# """
# <style>
# body {
#     background: linear-gradient(to right, #4e79a7, #86a8e7);
#     color: lightblue;
# }
# h1 {
#     color: #4e79a7;
# }
# </style>
# """,
# unsafe_allow_html=True
# )

# # Show sentiment output
# if predict_button and text:
#     sentiment, score = predict_sentiment(text)
#     if sentiment == "Positive":
#         st.success(f"The sentiment is {sentiment} with a score of {score*100:.2f}%!")
#     elif sentiment == "Negative":
#         st.error(f"The sentiment is {sentiment} with a score of {score*100:.2f}%!")
#     else:
#         st.warning(f"The sentiment is {sentiment} with a score of {score*100:.2f}%!")


import streamlit as st
import transformers

# Load model and tokenizer
model = transformers.AutoModelForSequenceClassification.from_pretrained("DeeeTeeee01/mytest_trainer_roberta-base")
tokenizer = transformers.AutoTokenizer.from_pretrained("DeeeTeeee01/mytest_trainer_roberta-base")

@st.cache_resource
def predict_sentiment(text):

  inputs = tokenizer(text, return_tensors="pt")

  outputs = model(**inputs)
  
  # # Get full model outputs 
  # outputs = model(text)
  
  # Extract probabilities
  negative = outputs[0][0]
  positive = outputs[0][1] 
  neutral = outputs[0][2]

  return negative, positive, neutral

# Page config
st.set_page_config(page_title="Sentiment Analysis", page_icon=":smile:") 

# Title and intro text
st.header("Predict Text Sentiment")
st.write("Enter text below to classify its sentiment as Positive, Negative or Neutral")

# Input text 
text = st.text_input("Enter text:")

# Predict button
predict_button = st.button("Predict") 

# Prediction output
if predict_button and text:
  
  # Get probabilities
  negative, positive, neutral = predict_sentiment(text)

  # Display probabilities
  st.metric("Negative", f"{negative*100:.2f}%")
  st.metric("Positive", f"{positive*100:.2f}%")
  st.metric("Neutral", f"{neutral*100:.2f}%")