Scoring system of each model

[AtomAnu]

@gbifvyihk Can you verify the following for me?

It would seem that the scoring method/function for regression models (e.g. Ridge Regression) is the score(X,y) method/function. This function. would return R^2 (coefficient of determination) regression score function. More info can be found here: https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.Ridge.html#sklearn.linear_model.Ridge.score.

However, for classification models, the method accuracy_score(y, y_pred) would return the fraction of correctly classified samples (float). More info can be found here: https://scikit-learn.org/stable/modules/generated/sklearn.metrics.accuracy_score.html.

My opinion is that we should compare regression models and classification models separately.

[AtomAnu]

Also, the method currently does not perform cross-validation to fine tune the hyperparameters. I intend to add it as well.

[sherrydodo]

I have asked the other team, and they say they are using predicted data to predict the test data, where they had different inputs with us. To make this more clear, they are predicting one match from the last fewer five and they predicted the first match as they are both the same, using average value.