Timeline for Log Loss function in scikit-learn returns different values
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 3, 2018 at 12:36 | vote | accept | GeneticsGuy | ||
| Jan 3, 2018 at 12:35 | comment | added | GeneticsGuy | My confusion came from how scikit learn handles encodings. It must use hashes to encode the labels. The ordering of hashed labels would the same regardless of the first samples label. To be specific, if the class of the first sample in the test set differs from that in the training set, the encoding of the classes remains the same, so the log loss of the test set and the train set will be comparable. Thanks for the help. | |
| Jan 3, 2018 at 12:27 | comment | added | Nikolas Rieble | That is an excellent example of an unjustified assumption. If you read the source code you will find that they use scikit-learn.org/stable/modules/generated/…. Then again have a look at this function and test it. | |
| Jan 3, 2018 at 12:19 | comment | added | GeneticsGuy |
Shouldn't both labels be interpreted as [0, 1, 1, 0]? log_loss is receiving the labels but converting them to two different binary encodings. I would've thought the first label encountered would be classed as 0, and the second as 1.
|
|
| Jan 3, 2018 at 12:05 | history | answered | Nikolas Rieble | CC BY-SA 3.0 |