NBARNN 1ST Attempt for WPI Hackathon 2024

Authors: Artem Frenk, Tair Kareneyev

Usage: Install required packages and csv file (or use provided get_data.py to get the csv yourself) then change FILE constant to local folder location of csv. If the csv is to be redownloaded, make sure the location is reflected in the code.

To adjust parameters of the data collected, go to lines 48 and 51 to adjust general collection and masking criteria respectively.

Example:

Here, this changes the code so that the end result csv contains NBA player data only from the years of 2003 to 2013.

Dependencies/Packages Needed:

Pandas: https://pandas.pydata.org/docs/index.html

Requests: https://requests.readthedocs.io/en/latest/

nba_api: https://github.com/swar/nba_api

tqdm: https://tqdm.github.io/

sklearn: https://github.com/scikit-learn/scikit-learn

tensorflow: https://github.com/tensorflow

keras: https://github.com/keras-team/keras

matplotlib: https://matplotlib.org/3.5.3/index.html

scipy: https://scipy.org/

copy: https://github.com/python/cpython/blob/3.12/Lib/copy.py

More info/sources:

Diederik P. Kingma, J. (2015). Adam: A Method for Stochastic Optimization. ICLR (Poster)

Tieleman, T. and Hinton, G. (2012). Lecture 6.5-rmsprop: Divide the Gradient by a Running Average of Its Recent Magnitude. COURSERA: Neural Networks for Machine Learning, 4, 26-31.

He, K., Zhang, X., Ren, S., & Sun, J. (2015). Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification. 2015 IEEE International Conference on Computer Vision (ICCV), 1026-1034.

Ioffe, S., & Szegedy, C. (2015). Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. ArXiv, abs/1502.03167.

Fukushima, K. Cognitron: A self-organizing multilayered neural network. Biol. Cybernetics 20, 121–136 (1975). https://doi.org/10.1007/BF00342633

Dixon, W. J. (1960). "Simplified Estimation from Censored Normal Samples". Annals of Mathematical Statistics. 31 (2): 385–391. doi:10.1214/aoms/1177705900.

Bridle, J.(1989). Training Stochastic Model Recognition Algorithms as Networks can Lead to Maximum Mutual Information Estimation of Parameters (NIPS)