stock.py

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import yfinance as yf
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error, r2_score
from xgboost import XGBRegressor
import streamlit as st
from testapidata import df

print(df)

print(df.isnull())

# Create previous 30 month features
for i in range(1, 31):
    df[f'prev_{i}'] = df['value'].shift(i)


# Features
X = df[[f'prev_{i}' for i in range(1, 31)]]
X = X.astype(float)

df['value'] = pd.to_numeric(df['value'], errors='coerce')

# Target
y = df['value']

# Train Test Split

X_train, X_test, y_train, y_test = train_test_split(
    X,
    y,
    test_size=0.2,
    shuffle=False
)


# XGBoost Model

model = XGBRegressor(
    n_estimators=300,
    learning_rate=0.05,
    max_depth=5,
    subsample=0.8,
    colsample_bytree=0.8,
    random_state=42
)

# Train Model

model.fit(X_train, y_train)

# Predictions

predictions = model.predict(X_test)

# Metrics

mse = mean_squared_error(y_test, predictions)
r2 = r2_score(y_test, predictions)

print("Mean Squared Error:", mse)
print("R2 Score:", r2)

# Show predictions

results = pd.DataFrame({
    'Actual': y_test.values,
    'Predicted': predictions
})

print(results)


# Function for future prediction
def predict_future():
    
    latest_values = df['value'].tail(30).values

    # Reverse values
    latest_values = latest_values[::-1]

    # Create input dataframe
    future_input = pd.DataFrame(
        [latest_values],
        columns=[f'prev_{i}' for i in range(1, 31)]
    )

    # Predict
    prediction = model.predict(future_input)

    return prediction[0]

# Plot graph

plt.figure(figsize=(12,6))
plt.plot(y_test.values, label='Actual')
plt.plot(predictions, label='Predicted')
plt.title('Actual vs Predicted Stock Values')
plt.xlabel('Time')
plt.ylabel('Stock Value')
plt.legend()
plt.grid(True)


plt.savefig("data.png")

# Predict next value

latest_data = X.tail(1)
next_prediction = model.predict(latest_data)

print("Next Predicted Value:", next_prediction[0])