import streamlit as st
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# Load the CSV file
@st.cache_data
def load_data():
    data = pd.read_csv('SME Survey.csv')
    return data

data = load_data()

# Title of the app
st.title('SME Survey Data Exploration')

# Display the raw data
st.subheader('Raw Data')
st.dataframe(data)

# Filters
st.sidebar.header('Filters')
business_nature = st.sidebar.multiselect('Select Business Nature', data['What is the nature of your business?'].unique(), default=data['What is the nature of your business?'].unique())
employee_range = st.sidebar.multiselect('Select Employee Range', data['How many employees does your business have?'].unique(), default=data['How many employees does your business have?'].unique())
familiar_with_ai = st.sidebar.multiselect('Familiarity with AI', data['Are you familiar with the concept of Artificial Intelligence (AI)?'].unique(), default=data['Are you familiar with the concept of Artificial Intelligence (AI)?'].unique())

# Apply filters
filtered_data = data[
    (data['What is the nature of your business?'].isin(business_nature)) &
    (data['How many employees does your business have?'].isin(employee_range)) &
    (data['Are you familiar with the concept of Artificial Intelligence (AI)?'].isin(familiar_with_ai))
]

# Display filtered data
st.subheader('Filtered Data')
st.dataframe(filtered_data)

# Analysis and Visualizations
st.subheader('Analysis and Visualizations')

# Bar chart for business nature
st.write('### Distribution of Business Nature')
fig, ax = plt.subplots(figsize=(10, 6))
sns.countplot(y=filtered_data['What is the nature of your business?'], order=filtered_data['What is the nature of your business?'].value_counts().index)
plt.title('Number of Businesses by Nature')
plt.xlabel('Count')
plt.ylabel('Business Nature')
st.pyplot(fig)

# Bar chart for employee range
st.write('### Distribution of Employee Range')
fig, ax = plt.subplots(figsize=(10, 6))
sns.countplot(y=filtered_data['How many employees does your business have?'], order=filtered_data['How many employees does your business have?'].value_counts().index)
plt.title('Number of Businesses by Employee Range')
plt.xlabel('Count')
plt.ylabel('Employee Range')
st.pyplot(fig)

# Pie chart for familiarity with AI
st.write('### Familiarity with AI')
fig, ax = plt.subplots(figsize=(8, 8))
filtered_data['Are you familiar with the concept of Artificial Intelligence (AI)?'].value_counts().plot(kind='pie', autopct='%1.1f%%', startangle=90)
plt.title('Familiarity with AI')
plt.ylabel('')
st.pyplot(fig)

# Bar chart for concerns about AI
st.write('### Concerns about Adopting AI')
fig, ax = plt.subplots(figsize=(10, 6))
sns.countplot(y=filtered_data['What concerns do you have about adopting AI in your business?'], order=filtered_data['What concerns do you have about adopting AI in your business?'].value_counts().index)
plt.title('Concerns about Adopting AI')
plt.xlabel('Count')
plt.ylabel('Concerns')
st.pyplot(fig)

# Bar chart for willingness to invest in AI
st.write('### Willingness to Invest in AI Solutions Annually')
fig, ax = plt.subplots(figsize=(10, 6))
sns.countplot(y=filtered_data['How much are you willing to invest in AI solutions annually?'], order=filtered_data['How much are you willing to invest in AI solutions annually?'].value_counts().index)
plt.title('Willingness to Invest in AI Solutions Annually')
plt.xlabel('Count')
plt.ylabel('Investment Range')
st.pyplot(fig)

# Bar chart for primary goal in adopting AI
st.write('### Primary Goal in Adopting AI')
fig, ax = plt.subplots(figsize=(10, 6))
sns.countplot(y=filtered_data['What would be your primary goal in adopting AI for your business?'], order=filtered_data['What would be your primary goal in adopting AI for your business?'].value_counts().index)
plt.title('Primary Goal in Adopting AI')
plt.xlabel('Count')
plt.ylabel('Primary Goal')
st.pyplot(fig)

# Comparison chart for business challenges by business nature
st.write('### Top Business Challenges by Business Nature')
challenges = filtered_data['Which business challenges do you face most often? (Rank top 3)'].str.split(';').explode().str.strip()
challenge_counts = challenges.value_counts()
challenge_df = filtered_data.merge(challenges.rename('Business Challenge'), left_index=True, right_index=True)
challenge_pivot = challenge_df.pivot_table(index='Business Challenge', columns='What is the nature of your business?', aggfunc='size', fill_value=0)

fig, ax = plt.subplots(figsize=(12, 8))
challenge_pivot.plot(kind='bar', ax=ax)
plt.title('Top Business Challenges by Business Nature')
plt.xlabel('Business Challenge')
plt.ylabel('Count')
plt.xticks(rotation=45)
st.pyplot(fig)

# Comparison chart for willingness to invest in AI by business nature
st.write('### Willingness to Invest in AI by Business Nature')
investment_df = filtered_data.copy()
investment_df['Investment Range'] = pd.Categorical(investment_df['How much are you willing to invest in AI solutions annually?'], categories=[
    'Below R 5000 per year',
    'Between R 6000 - R 10 000 per year',
    'Between R 20 000 - R 50 000 per year',
    'More than R50 000 per year'
], ordered=True)

investment_pivot = investment_df.pivot_table(index='Investment Range', columns='What is the nature of your business?', aggfunc='size', fill_value=0)

fig, ax = plt.subplots(figsize=(12, 8))
investment_pivot.plot(kind='bar', ax=ax)
plt.title('Willingness to Invest in AI by Business Nature')
plt.xlabel('Investment Range')
plt.ylabel('Count')
plt.xticks(rotation=45)
st.pyplot(fig)

# Comparison chart for primary goal in adopting AI by business nature
st.write('### Primary Goal in Adopting AI by Business Nature')
goal_df = filtered_data.copy()
goal_df['Primary Goal'] = goal_df['What would be your primary goal in adopting AI for your business?']

goal_pivot = goal_df.pivot_table(index='Primary Goal', columns='What is the nature of your business?', aggfunc='size', fill_value=0)

fig, ax = plt.subplots(figsize=(12, 8))
goal_pivot.plot(kind='bar', ax=ax)
plt.title('Primary Goal in Adopting AI by Business Nature')
plt.xlabel('Primary Goal')
plt.ylabel('Count')
plt.xticks(rotation=45)
st.pyplot(fig)

# Show the number of businesses using AI tools
st.write('### Number of Businesses Using AI Tools')
ai_usage = filtered_data['Have you used any AI tools in your business?'].value_counts()
st.bar_chart(ai_usage)

# Show the number of businesses interested in a free trial of AI tools
st.write('### Number of Businesses Interested in a Free Trial of AI Tools')
free_trial_interest = filtered_data['Would you be interested in a free trial of AI tools tailored to your business?'].value_counts()
st.bar_chart(free_trial_interest)

# Show the number of businesses knowing their sales revenue in real time
st.write('### Number of Businesses Knowing Sales Revenue in Real Time')
real_time_revenue = filtered_data['Do you know your sales revenue in real time?'].value_counts()
st.bar_chart(real_time_revenue)

# Show the preferred method of learning about new technologies
st.write('### Preferred Method of Learning about New Technologies')
learning_method = filtered_data['How do you prefer to learn about new technologies like AI?'].value_counts()
st.bar_chart(learning_method)

# Show the number of businesses by primary goal in adopting AI
st.write('### Number of Businesses by Primary Goal in Adopting AI')
primary_goal = filtered_data['What would be your primary goal in adopting AI for your business?'].value_counts()
st.bar_chart(primary_goal)

# Show the number of businesses by familiarity with AI
st.write('### Number of Businesses by Familiarity with AI')
familiarity_with_ai = filtered_data['Are you familiar with the concept of Artificial Intelligence (AI)?'].value_counts()
st.bar_chart(familiarity_with_ai)

# Show the number of businesses by concerns about AI
st.write('### Number of Businesses by Concerns about AI')
concerns_about_ai = filtered_data['What concerns do you have about adopting AI in your business?'].value_counts()
st.bar_chart(concerns_about_ai)

# Show the number of businesses by willingness to invest in AI
st.write('### Number of Businesses by Willingness to Invest in AI')
willingness_to_invest = filtered_data['How much are you willing to invest in AI solutions annually?'].value_counts()
st.bar_chart(willingness_to_invest)

# Show the number of businesses by current technological solutions
st.write('### Number of Businesses by Current Technological Solutions')
current_tech_solutions = filtered_data['What technological solutions do you currently use?'].str.split(';').explode().str.strip().value_counts()
st.bar_chart(current_tech_solutions)

# Show the number of businesses by needed technology solutions
st.write('### Number of Businesses by Needed Technology Solutions')
needed_tech_solutions = filtered_data['What technology solutions do you need for your business?'].str.split(';').explode().str.strip().value_counts()
st.bar_chart(needed_tech_solutions)

# Show the number of businesses by operational tasks handling
st.write('### Number of Businesses by Operational Tasks Handling')
operational_tasks = filtered_data['How do you currently handle the operational tasks of Customer support; Inventory management; Accounting/Finance and etc?'].str.split(';').explode().str.strip().value_counts()
st.bar_chart(operational_tasks)

# Show the number of businesses by business challenges
st.write('### Number of Businesses by Business Challenges')
business_challenges = filtered_data['Which business challenges do you face most often? (Rank top 3)'].str.split(';').explode().str.strip().value_counts()
st.bar_chart(business_challenges)