Simple time series prediction using LSTM implemented with Pytorch.
-
RNN_sine_wave - This notebook deals with the methodology of Time Series prediction for a sine wave, a manually created synthetic dataset, using LSTM. (Residual)
-
RNN_sales - This notebook deals with predicting sales of alcohol in the next year as a time series pattern. (Seasonal)
-
RNN_energy_production - This notebook deals with predicting production of energy in the next year as a time series pattern. (Seasonal + Linear)
Python 3 is required.
Follow the steps for installation given in the official website of Pytorch.
Install the required libraries using the pip package manager and requirements.txt file.
Install the required libraries using the pip package manager.
For pip version 19.1 or above:
pip install -r requirements.txt --user
or
pip3 install -r requirements.txt --user