cbalint13 opened a new pull request, #14468:
URL: https://github.com/apache/tvm/pull/14468

   This PR fix the latest xgboost >= 1.7.5 behaviour requiring binarized labels.
   
   * The strict behaviour was introduced by 
https://github.com/dmlc/xgboost/pull/8931
   * A possible solution (also 
[suggested](https://github.com/dmlc/xgboost/pull/9007#issuecomment-1494739265)) 
would be to binarize by threshold at the evaluation  calls.
   * This is a continuation on https://github.com/apache/tvm/pull/14036 xgboost
   
   This address both ```autotune``` and ```autoscheduler```.
   
   **Note:**
   Unsure about TVM overall tunner impact, but we can introduce more 
sophisticated way of measuring AP like [PASCAL evenly spaced 
one](https://en.wikipedia.org/wiki/Evaluation_measures_(information_retrieval)#Average_precision),
 the advantages are unclear and would require extensive comparative tests.
   
   ---
   The errors cought on TVM **autotune** process:
   ```
   File 
"/usr/lib64/python3.11/site-packages/tvm/autotvm/tuner/xgboost_cost_model.py"
   , line 538, in after_iteration
       bst_eval = model.eval_set(self.evals, epoch, feval)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   File "/usr/lib64/python3.11/site-packages/xgboost/core.py", line 1995, in 
eval_set
       _check_call(
   File "/usr/lib64/python3.11/site-packages/xgboost/core.py", line 270, in 
_check_call
       raise XGBoostError(py_str(_LIB.XGBGetLastError()))
   xgboost.core.XGBoostError: [15:59:59] 
/builddir/build/BUILD/xgboost/src/common/ranking_utils.h:378: 
   Check failed: is_binary: MAP can only be used with binary labels.
   ```
   ---
   
   Cc @Sunny-Island , @zxybazh , @junrushao , @vinx13 , please help with the 
review.
   
   Thanks,
   ~Cristian.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to