Romid
  • Member for 4 years, 8 months
  • Last seen more than a month ago
1 answers
2 votes
115 views
Oversampling possible improvement
1 votes

The evaluation of an imbalanced dataset should be done with more than one single metric since you need to evaluate the performance on both the majority and the minority classes. In the case that is ...

View answer
1 answers
1 votes
37 views
Filling created feature with values
0 votes

If I understood your question right - you've created a new feature age<55 - which gets True/False, hence, there is no need for the second feature since it has a perfect (negative) correlation with ...

View answer
4 answers
0 votes
491 views
What is the purpose of standardization in machine learning?
2 votes

There are several reasons for the standardization, the relevant reasons for the KNN algorithm important since the algorithm is based on calculating the distance between neighbours. Let's assume that ...

View answer
1 answers
2 votes
351 views
Filtering a panda dataframe in one line
Accepted answer
1 votes

The first groupby - counts the number of persons per day per meal The second groupby - counts the number of unique persons in each day The inner merge between the 2 - matches the number of persons ...

View answer
2 answers
1 votes
430 views
A robust metric in the presence of class imbalance
Accepted answer
3 votes

For evaluating the classification of a highly imbalanced, there are several measures that you may consider. Remember, that in such a problem we would prefer a measure that is not biased towards one of ...

View answer
1 answers
2 votes
1k views
How does SHAP values help us to determine importance of a feature for a model trained by gradient boost?
Accepted answer
0 votes

I think that the decision tree that appears in the second article is just illustrating the xgboost model that the shap is applied on. I would like to suggest you to read Christoph Molnar tutorial ...

View answer
2 answers
6 votes
177 views
Can we specify the number of data generated(minority class) using SMOTE?
3 votes

Some of the packages such the one in python (imbalanced-learn) allows you to set the balancing ratio (which is 1 in the case of 50% minority and 50% majorities). If you are not using a package ...

View answer
2 answers
2 votes
102 views
Explainability ML Methods
1 votes

You can look at the different evaluations as evaluation of the different stages of the model. by evaluating the model's performance on the training dataset, you could assess how and what the model ...

View answer
3 answers
1 votes
479 views
How to deal with highly skewed (on counts) dependent variables?
0 votes

You are generally right, but as you have mentioned these are important features and you would need to figure the way using them as they are with such a low signal below 2%. You may try building more ...

View answer
1 answers
3 votes
74 views
How to handle associated features in machine learning
1 votes

The RandomForest algorithm may capture such interaction by splitting an upper node by some condition on p1 and a lower node by another condition on p2, the link that you are expecting is captured if ...

View answer
1 answers
2 votes
205 views
n days back average for each day
Accepted answer
1 votes

You can use the pandas rolling method for that. back = 3 cols = ['high','low'] # calculate rolling mean on the dataframe's columns of interest df_back_average = df[cols].rolling(back).mean() # ...

View answer
2 answers
3 votes
6k views
Imbalanced dataset: how to deal with test data?
1 votes

The idea of balancing the training set + validating the balancing method is for being able to generalize your model that is would discriminate (in classification assignment) better a sample from the ...

View answer
1 answers
1 votes
602 views
How accurate does my machine learning model need to be?
4 votes

The accuracy level that should satisfy you depends on the reliability that your customer would have on it and the costs matrix of false predictions. He should put into consideration the costs of the ...

View answer