import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
import streamlit as st
import plotly.figure_factory as ff

st.title('A Simple Profit Prediction 💵')
st.image('https://assets2.ignimgs.com/2013/11/12/the-hobbit-bilbo-gold-1280jpg-e9649d_160w.jpg?width=1280')

st.write('### Data ')
df = pd.read_csv("https://raw.githubusercontent.com/amankharwal/Website-data/master/Startups.csv")
st.dataframe(df)
st.write('This data is about the R&D spend, Administration cost, Marketing Spend, State of operation, and the historical profit generated by 50 startups.')


df_graph = df.groupby('State')['Profit'].sum().reset_index()
st.write("### Total Profits by State")
st.bar_chart(df_graph, x="State", y="Profit")


numeric_columns = df.select_dtypes(include=['float64', 'int64'])
corr_matrix = numeric_columns.corr()
fig = ff.create_annotated_heatmap(z=corr_matrix.values,
                                  x=numeric_columns.columns.tolist(),
                                  y=numeric_columns.columns.tolist(),
                                  annotation_text=corr_matrix.round(2).values)
fig.update_layout()
st.write("### Correlation Heatmap")
st.plotly_chart(fig)

st.write('There is a clear correlation between R&D and Profit.Also, administration sucks.')

#model
st.write('### Model Training')
x = df[["R&D Spend", "Administration", "Marketing Spend"]]
y = df["Profit"]
x = x.to_numpy()
y = y.to_numpy()
y = y.reshape(-1, 1)
from sklearn.model_selection import train_test_split
xtrain, xtest, ytrain, ytest = train_test_split(x, y, test_size=0.2, random_state=42)
from sklearn.linear_model import LinearRegression
model = LinearRegression()
model.fit(xtrain, ytrain)
st.write(f"I after train test split, I have defined a LinerRegression Model. It's r2 score is. {model.score(xtest, ytest)}, meaning 90% accuracy")

st.write('### Using It')
st.write('Change the expenditures to see the difference in profit.')
# Input boxes for features
feature1 = st.number_input('R&D Spend', value=100000)
feature2 = st.number_input('Administration Spend', value=100000)
feature3 = st.number_input('Marketing Spend', value=100000)

features = np.array([[feature1, feature2, feature3]])

# Assuming 'model' is your trained model
prediction = model.predict(features)
st.write(f'Predicted Profit: {prediction[0]}')

st.write('### Conclusion')
st.write('This model abides the expenditure/profit relation of the 50 startups in this dataset. If you are one of them, this app will be useful. Otherwise, to make a more general model, we simply need more data.')