Bias vs Variance
Differences between High bias and High variance:
* High bias results in under fitting the training data. The below picture is evident that it is not able to predict or fit the training data well.
ex:
* High variance results in over fitting the data. The below picture tells that it fits the training data well, but not able generalize the data ie., it may not do well for new examples(over fitting generally makes a complex model).
Ex:
* A fair enough model might be like shown in the below figure:
Ex:
* High bias results in under fitting the training data. The below picture is evident that it is not able to predict or fit the training data well.
* High variance results in over fitting the data. The below picture tells that it fits the training data well, but not able generalize the data ie., it may not do well for new examples(over fitting generally makes a complex model).
Ex:
Ex:
Comments
Post a Comment