Hyperparameter Optimization Needed for Improved Accuracy and Efficiency in Reduced Redundancy Prediction Model

[agnij-dutta]

Is this a unique feature?

• [X] I have checked "open" AND "closed" issues and this is not a duplicate

Is your feature request related to a problem/unavailable functionality? Please describe.

The current implementation of the reduced redundancy prediction model lacks hyperparameter optimization, which hinders its performance and accuracy. The model's hyperparameters are not tuned to their optimal values, resulting in subpar predictions and reduced efficiency.

The absence of hyperparameter optimization leads to:

Suboptimal model performance, resulting in reduced accuracy and recall Inefficient use of model resources, leading to increased computational time and costs Limited understanding of the model's performance, making it challenging to identify areas for improvement

Proposed Solution

To address this issue, I propose implementing hyperparameter optimization techniques, such as GridSearchCV, to tune the model's hyperparameters to their optimal values. This will enable the model to achieve improved accuracy and recall, while also enhancing its efficiency.

Expected Benefits:

The implementation of hyperparameter optimization is expected to bring the following benefits:

Improved accuracy and recall of the reduced redundancy prediction model Enhanced efficiency, resulting in reduced computational time and costs More informative evaluation metrics, providing a comprehensive understanding of the model's performance

Related Pull Request:

This issue is addressed in the pull request [PR #121], which introduces several improvements to the hyperparameter tuning and model evaluation process. The changes include:

Polynomial feature transformation to create new features that can help improve the model's performance Voting regressor to combine the predictions of multiple models and produce a more accurate prediction Hyperparameter tuning using GridSearchCV to optimize the performance of the voting regressor Calculation of additional evaluation metrics, including accuracy, precision, recall, F1 score, and confusion matrix Error handling to fix an invalid parameter in the VotingRegressor class

Screenshots

No response

Do you want to work on this issue?

Yes

If "yes" to above, please explain how you would technically implement this (issue will not be assigned if this is skipped)

To address the issue of lacking hyperparameter optimization in the reduced redundancy prediction model, I propose the following technical implementation:

Step 1: Polynomial Feature Transformation

Utilize the PolynomialFeatures class from the sklearn.preprocessing library to create a polynomial feature transformer. Implement the create_poly_features function to create and fit the polynomial feature transformer to the training data. Step 2: Voting Regressor

Employ the VotingRegressor class from the sklearn.ensemble library to combine the predictions of multiple models. Introduce the create_voting_regressor function to create and fit the voting regressor to the training data. Step 3: Hyperparameter Tuning

Leverage the GridSearchCV class from the sklearn.model_selection library to perform hyperparameter tuning for the voting regressor. Modify the GridSearchCV call to tune the weights parameter of the VotingRegressor class. Step 4: Evaluation Metrics

Calculate additional evaluation metrics, including accuracy, precision, recall, F1 score, and confusion matrix, using the sklearn.metrics library. Implement the necessary code to compute and display these metrics for the voting regressor. Step 5: Error Handling

Fix the error caused by an invalid parameter in the VotingRegressor class by removing the invalid voting parameter from the GridSearchCV call. Existing Code and Libraries

The implementation will build upon the existing codebase and utilize the following libraries: sklearn.preprocessing for polynomial feature transformation sklearn.ensemble for voting regressor sklearn.model_selection for hyperparameter tuning sklearn.metrics for evaluation metrics

[github-actions[bot]]

Ensure the issue is not similar or previously being worked on.Thanks for your time

[agnij-dutta]

I have checked among all the open and closed issues... it has been done for the targeted models... but no changes have been made to reduced redundancy models... hence I added hyperparameter optimization. also Voting Regressor was not raised as an issue or PR... please add hacktoberfest as a label.

[agnij-dutta]

please check [PR #121} for the changes I made for this issue. if found ok, please close both this issue and pull request. Please add the hacktoberfest as a label in that PR as well.