RNN-Google-Stock-Price

Google Stock Price Prediction with LSTM This project demonstrates how to build and train a Long Short-Term Memory (LSTM) network for predicting Google stock prices. The model utilizes historical stock price data to forecast future prices. The dataset includes training and test files with historical stock prices for Google. The model is implemented using TensorFlow/Keras and consists of four LSTM layers with 50 units each, dropout layers to prevent overfitting, and a Dense layer for the final output. The training process involves 100 epochs with a batch size of 32, using Mean Squared Error as the loss function and Adam as the optimizer. Evaluation is performed using Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) on the test dataset. The results are visualized through plots comparing the real and predicted stock prices, as well as the training and validation loss over epochs. To run the script, ensure that the data files are placed in the /content/ directory and execute the script in a Python environment with the necessary packages installed, using the command python main.py. The project is licensed under the MIT License. For more details, refer to the LICENSE file.