Getting error while running this code.

[ashwinirahangdale]

I am executing this code for analyzing the loss function for learning to rank. But while running this mlr I have got numbers of errors. I am attaching a snapshot of the error in attachment.

[bmcfee]

(Side note: it's more helpful to copy-paste the text of the error, rather than a screenshot. It's easier to read for me, and it makes it possible for other people to find this thread by searching for the error message.)

It's hard to know what's going on there -- it looks like the MAP separation oracle is not returning the right number of values, but I don't see how this can be the case. Do you get the same error message if you use auc or prec@k loss?

[ashwinirahangdale]

Respected sir, I have tried to run this code with LOSS='AUC' and 'Prec@k' but getting the same error.

Error with Loss= 'AUC' Loading Wine data z-scoring features Generating a 80/20 training/test split Training with C=1.00e-02, Delta=mAP Undefined function or variable 'initializeFullMKL'.

Error in mlr_train (line 298) W = INIT(X);

Error in mlr_demo (line 24) [W, X, Diagnostics] = mlr_train(Xtrain, Ytrain, C, 'AUC');

Error with Loss= 'MAP' mlr_demo Loading Wine data z-scoring features Generating a 80/20 training/test split Training with C=1.00e-02, Delta=mAP Undefined function or variable 'initializeFullMKL'.

Error in mlr_train (line 298) W = INIT(X);

Error in mlr_demo (line 24) [W, X, Diagnostics] = mlr_train(Xtrain, Ytrain, C, 'MAP');

‌ https://mailtrack.io/ Sent with Mailtrack https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality&

ASHWINI RAHANGDALE Ph.D. Research Scholar VNIT Nagpur

On Wed, Feb 28, 2018 at 7:37 PM, Brian McFee notifications@github.com wrote:

(Side note: it's more helpful to copy-paste the text of the error, rather than a screenshot. It's easier to read for me, and it makes it possible for other people to find this thread by searching for the error message.)

It's hard to know what's going on there -- it looks like the MAP separation oracle is not returning the right number of values, but I don't see how this can be the case. Do you get the same error message if you use auc or prec@k loss?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/bmcfee/mlr/issues/5#issuecomment-369249656, or mute the thread https://github.com/notifications/unsubscribe-auth/AGU9YW7yCCJljXqL8antZAX9i2C0_WIxks5tZV2pgaJpZM4STuVa .

[bmcfee]

Undefined function or variable 'initializeFullMKL'.

This error indicates that the mlr subroutines are not in your path. Did you remember to addpath(genpath('/path/to/mlr'))?

[ashwinirahangdale]

Thank you for your previous suggestion. It is working but stuck at next stage. The errors are as follows

mlr_demo Loading Wine data z-scoring features Generating a 80/20 training/test split Training with C=1.00e-02, Delta=mAP Error using * Inner matrix dimensions must agree.

Error in structKernelMKL (line 3) H = Psi1(:)' * Psi2(:);

Error in mlr_train>expandRegularizer (line 461) Q(m+1,1) = STRUCTKERNEL(REG(W,K,1), PsiR{m+1});

Error in mlr_train (line 391) Q = expandRegularizer(Q, X, W);

Error in mlr_demo (line 24) [W, X, Diagnostics] = mlr_train(Xtrain, Ytrain, C, 'prec@k');

‌ https://mailtrack.io/ Sent with Mailtrack https://mailtrack.io?utm_source=gmail&utm_medium=signature&utm_campaign=signaturevirality&

ASHWINI RAHANGDALE Ph.D. Research Scholar VNIT Nagpur

On Fri, Mar 9, 2018 at 6:16 AM, Brian McFee notifications@github.com wrote:

Undefined function or variable 'initializeFullMKL'.

This error indicates that the mlr subroutines are not in your path. Did you remember to addpath(genpath('/path/to/mlr'))?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/bmcfee/mlr/issues/5#issuecomment-371675455, or mute the thread https://github.com/notifications/unsubscribe-auth/AGU9YSQ7Y84CbRB6vWvlAxvhweQyczqTks5tcdDKgaJpZM4STuVa .