SFIIPrediction is an open-source database designed for making binary winner predictions in the Street Fighter II video game. It operates by using data extracted from the percentage change in the health bar every 5 frames. To utilize the algorithm, download the dataset "Ultimate_sheets.xlsx", then open "runlstm.py" and execute it. You can modify the "round_interv" variable to any value between 0 and 1. This adjustment specifies the round progression percentage you wish to use as test data for making predictions.
# Set random seed for reproducibility
seed = 42
round_interv = 0.75
random.seed(a=seed)
np.random.seed(seed=seed)
torch.manual_seed(seed=seed)
# Set compute device to GPU if available or CPU otherwise
device = ("cuda" if torch.cuda.is_available() else "cpu")
Author Kittimate Chulajata, Sean Wu