This is a work in progress, forecasted data and models have not been tested for accuracy. Intended for educational purposes.
-
Run the setup.py via the command:
python3 setup.py install --user
-
Check to see if you have gpu support enabled by running the check_gpu.py, will return true/false:
python3 check_gpu.py --user
-
Add items to the items_to_predict.csv, if you are editing it via a text editor please be sure to add 4 commas:
item name,,,, -
Run the pytorch_predictions.py file:
python3 pytorch_predictions.py
-
Images will be generated and placed in an img folder located in the root project directory. This directory includes sub folders for all prices, training, and forecast images.
- https://pytorch.org/docs/stable/generated/torch.nn.LSTM.html
- https://www.kaggle.com/rodsaldanha/stock-prediction-pytorch
- https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.xticks.html
- https://stackabuse.com/time-series-prediction-using-lstm-with-pytorch-in-python/
- https://github.com/chriskok/GEPrediction-OSRS
- https://github.com/JonasHogman/osrs-prices-api-wrapper
- https://gist.github.com/adoskk/c3d96e4c7ae15a48c2a9ea8bc835ca39