Может ли кто-нибудь помочь мне сделать этот график более аккуратным? ⇐ Python

[Anonymous]

Код: Выделить всё

import math
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from sklearn import metrics
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
import datetime

# Load the Tesla Dataset
tesla_stock = pd.read_csv('C:/Users/Admin/Downloads/AI/Tesla.csv')

# Dynamically handle missing or renamed 'Close' column
possible_target_columns = ['Close', 'Last', 'Price', 'Adjusted Close', 'VWAP',
'Close/Last']
target_column = None

# Print available columns for debugging
print("Columns available in the dataset:", tesla_stock.columns)

for col in possible_target_columns:
if col in tesla_stock.columns:
target_column = col
print(f"Using '{col}' as the target column for predictions.")
break

# Handle case when no suitable column is found
if target_column is None:
print("No suitable column for predictions found. Please check the dataset.")
raise KeyError("Ensure the dataset includes a column with stock prices (e.g., 'Close',
'Close/Last').")

# Clean the numeric columns (removing '$' and converting to float)
for column in ['Open', 'High', 'Low', 'Close/Last', 'Volume']:
if column in tesla_stock.columns:
tesla_stock[column] = tesla_stock[column].replace('[\$,]', '',
regex=True).astype(float)

# Convert 'Date' to datetime and set as index
tesla_stock['Date'] = pd.to_datetime(tesla_stock['Date'])
tesla_stock.set_index('Date', inplace=True)

# Define features and target variable
features = ['Open', 'High', 'Low', 'Volume']
X = tesla_stock[features]
y = tesla_stock[target_column]

# Split data into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)

# Train the Linear Regression model
model = LinearRegression()
model.fit(X_train, y_train)

# Make predictions
predicted = model.predict(X_test)

# Evaluate the model
print("Model Score (R²):", model.score(X_test, y_test))
print("Mean Absolute Error:", metrics.mean_absolute_error(y_test, predicted))
print("Mean Squared Error:", metrics.mean_squared_error(y_test, predicted))
print("Root Mean Squared Error:", math.sqrt(metrics.mean_squared_error(y_test,
predicted)))

# Plot Actual vs Predicted Prices for the test set
dfr = pd.DataFrame({'Actual': y_test, 'Predicted': predicted})
plt.figure(figsize=(14, 8))
dfr.head(25).plot(kind='bar', figsize=(14, 8))
plt.title("Actual vs Predicted Prices")
plt.xlabel("Sample")
plt.ylabel("Price (USD)")
plt.xticks(rotation=45, ha="right")
plt.tight_layout()
plt.show()

# Predict the next 30 days
last_date = tesla_stock.index[-1]  # Last date in historical data
last_price = tesla_stock[target_column].iloc[-1]  # Last price in historical data
future_dates = [last_date + datetime.timedelta(days=i) for i in range(1, 31)]  #
Generate next 30 days

# Create a placeholder DataFrame for future features
future_features = pd.DataFrame(index=future_dates)
for feature in features:
if feature in tesla_stock.columns:
future_features[feature] = tesla_stock[feature].mean()  # Use the mean value of
each feature

# Predict future prices using the trained model
future_predictions = model.predict(future_features)

# Combine the historical and future data for a seamless plot
all_dates = list(tesla_stock.index) + list(future_dates)  # Combine historical and
future dates
all_prices = list(tesla_stock[target_column]) + list(future_predictions)  # Combine
historical and predicted prices

# Plot historical data and future predictions
plt.figure(figsize=(14, 8))
plt.plot(tesla_stock.index, tesla_stock[target_column], label='Historical Prices',
color='blue')
plt.plot(future_dates, future_predictions, label='30-Day Future Predictions',
color='red')
plt.title('Tesla Stock Price Prediction for the Next 30 Days')
plt.xlabel('Date')
plt.ylabel('Price (USD)')
plt.legend()
plt.xticks(rotation=45)
plt.show()

# Display future predictions
future_features['Predicted_Close'] = future_predictions
print(future_features[['Predicted_Close']])

введите здесь описание изображения
я пытаюсь, чтобы красная линия соединялась с концом графика и как бы изгибалась вниз до того места, где она находится. Вместо случайной короткой красной линии в начале, потому что это кажется немного беспорядочным. Я пытался это сделать, но у меня действительно были проблемы с этим. был бы очень признателен, если бы кто-нибудь мог мне помочь с этим, спасибо.

Подробнее здесь: https://stackoverflow.com/questions/792 ... bit-neater