dsckgec / learn-machine-learn Goto Github PK

Is your feature request related to a problem? Please describe.
The Support Vector Regression block has been left incomplete. This feature will further help us to implement another regression model with which we can predict the model with more accuracy.

Describe the solution you'd like
Implement Support Vector Regression on the given dataset given with the required preprocessed entities

Is your feature request related to a problem? Please describe.
The Random Forest Regression code is not predicting results with much accuracy. This feature will further help us to implement another regression model with which we can predict the model with more accuracy.

Describe the solution you'd like
Improve the accuracy of the Random Forest Regression on the given dataset given with the required preprocessed entities

Is your feature request related to a problem? Please describe.
The polynomial linear regression block has been left incomplete. This feature will further help us to implement another regression model with which we can predict the model with more accuracy.

Describe the solution you'd like
Implement Polynomial Regression on the given dataset given with the required preprocessed entities

The k-nearest neighbors algorithm, also known as KNN or k-NN, is a non-parametric, supervised learning classifier, which uses proximity to make classifications or predictions about the grouping of an individual data point.

###Requested Feature

Build a KNN classifier and in the output, f1-score, accuracy and confusion matrix must be printed, there is a function named metrics for printing the accuracy and confusion matrix.
Hyperparameter tuning can be done to improve the accuracy. As the dataset is imbalanced, do prefer f1 score as metric while training.

A Naive Bayes model is already present in the classification notebook.

Confusion Matrix
True positive (TP): Prediction is +ve and X is Cancerous, we want that
True negative (TN): Prediction is -ve and X is healthy, we want that too
False positive (FP): Prediction is +ve and X is healthy, false alarm, bad
False negative (FN): Prediction is -ve and X is Cancerous, the worst

Requested Feature

improvement in model prediction ( less False positive and less false negative(need more attention and improvement))
Cross validation or hyper parameter tuning can be done to improve the model.

About metrics

Choose Recall if the idea of false positives is far better than false negatives, in other words, if the occurrence of false negatives is unaccepted/intolerable, that you’d rather get some extra false positives(false alarms) over saving some false negatives, like in our cancerous example.

Is your feature request related to a problem? Please describe.
The R_2 scores block has been left incomplete. This feature will further help us to visualize the accuracies of all the models at a glance and will help us to pick up the better fitted algorithm on the dataset.

Describe the solution you'd like
Implement R_2 scores on the given dataset given with the predicted model values

###Requested Feature

Build a Feed-forward-neural-network, compile and fit on data and in the output, f1-score, accuracy and confusion matrix must be printed, there is a function named metrics for printing the accuracy and confusion matrix. plot accuracy and loss curve for train data and validation data
Hyperparameter tuning or scalling can be done to improve the accuracy. As the dataset is imbalanced, prefer f1 score as metric while training.
Avoid under fitting and over fitting of the model.

The Ridge Classifier, based on Ridge regression method, converts the label data into [-1, 1] and solves the problem with regression method. The highest value in prediction is accepted as a target class.

Requested Feature

Build a Ridge classifier and in the output, f1-score, accuracy and confusion matrix must be printed, there is a function named metrics for printing the accuracy and confusion matrix.
Hyperparameter tuning can be done to improve the accuracy. As the dataset is imbalanced, do prefer f1 score as metric while training.

###Requested Feature

Build a Logistic Regression model and in the output, f1-score, accuracy and confusion matrix must be printed, there is a function named metrics for printing the accuracy and confusion matrix.
Hyperparameter tuning can be done to improve the accuracy. As the dataset is imbalanced, do prefer f1 score as metric while training.

Is your feature request related to a problem? Please describe.
The Decision Tree Regression block has been left incomplete. This feature will further help us to implement another regression model with which we can predict the model with more accuracy.

Describe the solution you'd like
Implement Decision Tree Regression on the given dataset given with the required preprocessed entities

Requested Feature

create a python function which will take y_true and y_pred as arguments and output will look like the following picture -

A random forest is a meta estimator that fits a number of decision tree classifiers on various sub-samples of the dataset and uses averaging to improve the predictive accuracy and control over-fitting.

Requested Feature

Build a Random Forest classifier and in the output, f1-score, accuracy and confusion matrix must be printed, there is a function named metrics for printing the accuracy and confusion matrix.
Hyperparameter tuning can be done to improve the accuracy. As the dataset is imbalanced, do prefer f1 score as metric while training.