Overview:
This Python script uses a Decision Tree Classifier to predict stock volume based on historical data. The dataset is loaded from a CSV file (data.csv) containing stock information, and the machine learning model is trained to predict stock volumes. The predictions are then compared with the actual values, and the results are visualized using a bar chart.
Dependencies:
Make sure you have the following Python libraries installed:
pandas scikit-learn joblib graphviz yfinance matplotlib You can install the required dependencies using the following:
pip install pandas scikit-learn joblib graphviz yfinance matplotlib Usage:
#Clone the repository:
git clone https://github.com/blueraymusic/Prediction.git Install dependencies: pip install -r requirements.txt Run the script: python Prediction.ipynb
Code Explanation:
Data Loading: The script reads stock data from a CSV file (data.csv) using the pandas library. python Copy code file = 'data.csv' data = pd.read_csv(file) Data Preprocessing: The 'date' column is converted to a numeric format. python Copy code data['date'] = data['date'].apply(lambda x: float(x.split()[0].replace('-', ''))) Model Training: The Decision Tree Classifier is used to train the model on the stock data. python Copy code x_values = data.drop(columns=['volume', 'Name', 'date']) y = data['volume'] x_train, x_test, y_train, y_test = train_test_split(x_values, y, test_size=0.0005) model = DecisionTreeClassifier() model.fit(x_train, y_train) Prediction and Evaluation: The model predicts stock volumes and calculates the accuracy score. python Copy code predictions = model.predict(x_test) error = 0.10 * predictions score = accuracy_score(y_test, predictions) Visualization: The results are visualized using a bar chart. python Copy code results_df = pd.DataFrame({ 'Real Numbers': y_test, 'Predictions': predictions - error })
try: # Visualization using matplotlib ax = results_df.plot(kind='bar', figsize=(10, 6), legend=True) plt.title('Real Numbers vs Predictions') plt.xlabel('Volumes') plt.ylabel('Values') plt.xticks([]) # Remove x-axis tick labels plt.legend(loc='upper right') ax.yaxis.set_major_formatter(plt.FuncFormatter(lambda x, _: '{:.2f}'.format(x))) plt.show() except: print('<-----Could not Visualize the data----->') Model Saving and Loading: The trained model can be saved using joblib and loaded for future use.
- python
Save the trained model joblib.dump(model, 'model_trained.joblib')
Load the model model = joblib.load('model_trained.joblib') Stock Data Retrieval: Historical stock data is downloaded using the yfinance library and plotted. python Copy code Get historical stock data data = yf.download(ticker, '1980-12-12', '2023-11-17')
Plot adjusted close price data data['Volume'].plot() plt.show()