DOTA MATCH ANALYZER

Quick Introduction:

This project:

• Scrapes the Dotabuff to build a dataset
• Uses various Machine Learning/Deep Learning technologies to predict the outcome of the given matches.

Required Libraries:

• BeautifulSoup4
• lxml
• Tensorflow
• Pandas
• Scikit-Learn
• Numpy

DISCLAIMER:

Tensorflow doesn't let you use the most recent version of which is currently python3.11 . You can change the python version from the beginning of my files.

#!/usr/bin/env python3

Explaining the scripts:

classifiers:

• The first few lines import various libraries that are needed for the script, including os, tensorflow, pandas, and sklearn.
• The script splits the data and results datasets into training and testing sets using the train_test_split function from sklearn.model_selection.
• The script then imports the RandomForestClassifier class from sklearn.ensemble and creates an instance of this class called rfc. The rfc model is then fit to the training data using the fit method.
• The script uses the predict method of the rfc model to generate predictions for both the test data and the test_data dataset. The accuracy_score, function from sklearn.metrics is then used to calculate the accuracy of the model's predictions on both the test data and the test_data dataset.
• The script repeats this process for two other machine learning models: a decision tree classifier and a support vector classifier. The script imports the necessary classes, creates instances of the models, fits them to the training data, generates predictions, and calculates the accuracy of the predictions.

dense-neural-network:

• The first few lines import various libraries that are needed for the script, including os, tensorflow, pandas, and sklearn.
• The script sets some environment variables related to TensorFlow, which is a library for machine learning.
• The script loads a dataset from a CSV file using pandas.read_csv and processes the dataset by removing duplicates, dropping rows with missing values, and removing a column called 'Unnamed: 0'.
• The script creates two variables, X and y, to store the feature data and target data from the dataset, respectively. The X variable stores the values of all columns except for the 'Result' column, and the y variable stores the values of the 'Result' column.
• The script loads another dataset called test_data from a CSV file and processes it in a similar way as the first dataset. It also creates a test_results variable to store the values of the 'Result' column from the test_data dataset.
• The script creates an instance of the MinMaxScaler class from sklearn.preprocessing and uses it to scale the values in the X variable.
• The script then splits the scaled X and y datasets into training and testing sets using the train_test_split function from sklearn.model_selection.
• The script creates a neural network model using the Sequential class from tensorflow.keras. The model consists of several fully connected (Dense) layers with various activation functions and regularization techniques applied.
• The script compiles the model using the compile method, specifying the optimizer, loss function, and metrics to be used during training.
• The script creates an EarlyStopping callback object and passes it to the fit method of the model to interrupt training if the validation loss does not improve after a certain number of epochs.
• The script trains the model on the training data using the fit method, specifying the number of epochs, batch size, and validation split to be used.
• The script evaluates the model on the test data using the evaluate method and prints the test accuracy.
• The script generates predictions for the test data and the test_data dataset using the predict method of the model.

five-fold-neural-network

• The first few lines import various libraries that are needed for the script, including os, tensorflow, pandas, and sklearn.
• The script sets some environment variables related to TensorFlow, which is a library for machine learning.
• The script loads a dataset from a CSV file using pandas.read_csv and processes the dataset by removing duplicates, dropping rows with missing values, and removing a column called 'Unnamed: 0'.
• The script creates two variables, X and y, to store the feature data and target data from the dataset, respectively. The X variable stores the values of all columns except for the 'Result' column, and the y variable stores the values of the 'Result' column.
• The script loads another dataset called test_data from a CSV file and processes it in a similar way as the first dataset. It also creates a test_results variable to store the values of the 'Result' column from the test_data dataset.
• The script creates an instance of the MinMaxScaler class from sklearn.preprocessing and uses it to scale the values in the X variable.
• The script creates a KFold object from sklearn.model_selection with a specified number of splits and shuffle options.
• The script enters a loop to iterate over the splits generated by the KFold object. On each iteration, the script splits the scaled X and y datasets into training and testing sets using the train_index and test_index variables.
• The script creates neural network model using the Sequential class from tensorflow.keras. The model consists of several fully connected (Dense) layers with various activation functions and regularization techniques applied.
• The script compiles the model using the compile method, specifying the optimizer, loss function, and metrics to be used during training.
• The script creates an EarlyStopping callback object and passes it to the fit method of the model to interrupt training if the validation loss does not improve after a certain number of epochs.
• The script trains the model on the training data using the fit method, specifying the number of epochs, batch size, and validation split to be used.
• The script evaluates the model on the test data using the evaluate method and prints the test loss and accuracy.
• The script generates predictions for the test data and the test_data dataset using the predict method of the model.
• The script converts the predictions to binary values using a threshold of 0.5 and calculates the accuracy of the predictions on the test_results dataset using the accuracy_score function from sklearn.metrics.

NOTE: Datasets can be updated by using the fetch-script