import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

from transformers import pipeline

def sentiment_analysis1(text):
  # Load the sentiment analysis pipeline with the BERT model
  classifier = pipeline("sentiment-analysis", model="nlptown/bert-base-multilingual-uncased-sentiment")

  # Perform sentiment analysis
  result = classifier(text)
  return result

# Example usage:
financial_news = "The stock market experienced a significant downturn today amid concerns over inflation."
result = sentiment_analysis1(financial_news)
print("Sentiment Analysis Result:", result)


data = {
    'headline': [
         "Apple's revenue exceeds expectations with a strong quarterly report",
         "Major banks face scrutiny as financial regulations tighten",
         "Tesla's new factory investment raises concerns over debt levels"
     ]
}

df = pd.DataFrame(data)
df

df['sentiment'] = sentiment_analysis1(df['headline'].tolist())
print(df)

from transformers import pipeline

# Initialize the sentiment analysis model
model_name = "yiyanghkust/finbert-tone"
finbert = pipeline("sentiment-analysis", model=model_name)

# Function to analyze sentiment
def analyze_sentiment(text):
    results = finbert(text)
    return results

# Example usage:
financial_news = "The company's revenue has exceeded expectations, but concerns about regulatory challenges remain."
sentiment_result = analyze_sentiment(financial_news)

# Print the results
print("Sentiment Analysis Result:", sentiment_result)


df['sentiment'] = analyze_sentiment(df['headline'].tolist())
print(df)





pos="In two years at Texas State Dr. Yi was perhaps the most exceptional teacher \
I have had. His passion for what he teaches as well as his students well-being is \
evident. He went far beyond what is necessary to provide us with material via his canvas pages. \
I would highly highly recommend other students in Finance take his course(s) they would be better for it. \
Thank you"

neg ="Was hard to understand what the professor was saying. \
Materials on the test sometimes would reflect only a small fraction of the homework and review,\
the rest seemed to be vague in where he got the information from. \
It was also hard to stay engaged during class, I fell asleep every time I attended class. \
The lectures were practically useless to me"

#!pip install textblob

from textblob import TextBlob
print(TextBlob(pos).sentiment)
print(TextBlob(neg).sentiment)

!pip install vaderSentiment

from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer
vs_pos = SentimentIntensityAnalyzer().polarity_scores(pos)
vs_neg = SentimentIntensityAnalyzer().polarity_scores(neg)

print("{:-<65} {}".format(pos, str(vs_pos)))
print('--------------')
print("{:-<65} {}".format(neg, str(vs_neg)))
print(vs_pos)
print(vs_neg)

!pip install flair

from flair.models import TextClassifier
from flair.data import Sentence

classifier = TextClassifier.load('en-sentiment')

sentence = Sentence(pos)
classifier.predict(sentence)

# print sentence with predicted labels
print('Sentence above is: ', sentence.labels)

sentence = Sentence(neg)
classifier.predict(sentence)

# print sentence with predicted labels
print('Sentence above is: ', sentence.labels)



from transformers import pipeline

def sentiment_analysis1(text):
  # Load the sentiment analysis pipeline with the BERT model
  classifier = pipeline("sentiment-analysis", model="nlptown/bert-base-multilingual-uncased-sentiment")

  # Perform sentiment analysis
  result = classifier(text)
  return result

# Example usage:

result = sentiment_analysis1(pos)
print("Sentiment Analysis Result:", result)

result = sentiment_analysis1(neg)
print("Sentiment Analysis Result:", result)




from transformers import pipeline

# Initialize the sentiment analysis model
model_name = "yiyanghkust/finbert-tone"
finbert = pipeline("sentiment-analysis", model=model_name)

# Function to analyze sentiment
def analyze_sentiment(text):
    results = finbert(text)
    return results

# Example usage:
sentiment_result = analyze_sentiment(pos)

# Print the results
print("Sentiment Analysis Result:", sentiment_result)


# Example usage:
sentiment_result = analyze_sentiment(neg)

# Print the results
print("Sentiment Analysis Result:", sentiment_result)