ML Practitioner Assessement : Models score equal to 1

[Adrick]

Hi there,

I try to pass the ML Practitioner Certificate. I follow the Hands-on Instructions, but after splitting my original dataset into two smaller ones (train_data and validation_data), and after creating a baseline ML model applied to the train_data, I find out that both models (random forest and logistic regression) show a score of 1.

I wonder how this is possible. I retried multiple times, still the same. I followed every step mentioned previously. The only data preprocessing I did is to change the type of the columns of the school_data as mentioned in the previous steps.

Any idea ?

Thanks in advance.

[phh]

Hi @Adrick 

Did you include the grade column as a feature?

Cause the repeated (target variable)column is computed based on grade, the grade column should be rejected.

[Adrick]

Thanks for your response @phh .

That makes sense. I disabled the grade feature and tried again, and the score is lower (around 0.9). But the thing is, as the Hands-on Instructions suggest, I should face some diagnostic issues. That's not the case here. I have a score of 0.922, with no diagnostic issue at all.

I feel like something is wrong.