This repository contains Python code for analyzing and modeling CS:GO match data. The code focuses on extracting meaningful features from CS:GO demo files, training a model with and without word embeddings, and evaluating its performance in predicting win probability.
Make sure you have the following Python libraries installed:
- pandas
- glob
- xgboost
- scikit-learn
- os
- gensim
You can install them using the following command:
pip install pandas glob2 xgboost scikit-learn gensimDataset Collection Before running the code, make sure to set the correct path for your CS:GO demo files. Update the dir_name variable with the location of your stored CS:GO demo files (download the sample ones in this repository). A sample of 50 demos is included in this repository. For more demos visit https://docs.pureskill.gg/datascience/adx/csgo/csds/spec/#round_end---single_event
Replace this directory with the location of which your demos are stored as files containing parquet files.
dir_name = "C:\\Users\\kAMAL\\Desktop\\pureskill\\Demos"Important links if needed by reader:
https://docs.pureskill.gg/datascience/adx/csgo/csds/spec
https://docs.google.com/spreadsheets/d/11tDzUNBq9zIX6_9Rel__fdAUezAQzSnh5AVYzCP060c/htmlview