import streamlit as st
import pandas as pd
import numpy as np
import plotly.graph_objs as go
from keras.preprocessing.text import Tokenizer

# Set up the Streamlit app
st.set_page_config(page_title='Keras and Plotly Example')
st.sidebar.title('Word Frequency')

# Generate random collections of words and save to CSV file
def generate_data():
    data = []
    for i in range(5):
        words = np.random.choice(['apple', 'banana', 'cherry', 'date', 'elderberry'], size=100)
        data.append(' '.join(words))
    df = pd.DataFrame({'text': data})
    df.to_csv('word_data.csv', index=False)

# Load data from CSV file
def load_data():
    df = pd.read_csv('word_data.csv')
    return df

# Create a bar chart of word frequency
def plot_word_frequency(text):
    tokenizer = Tokenizer()
    tokenizer.fit_on_texts(text)
    word_counts = tokenizer.word_counts
    words = list(word_counts.keys())
    counts = list(word_counts.values())
    fig = go.Figure([go.Bar(x=words, y=counts)])
    fig.update_layout(title='Word Frequency')
    st.plotly_chart(fig)

# Main Streamlit app
if st.sidebar.button('Generate Data'):
    generate_data()
    st.sidebar.write('Data generated')
else:
    st.sidebar.write('Click "Generate Data" to generate new data')

df = load_data()
text = df['text'].tolist()
st.write(df)
plot_word_frequency(text)