[GitHub] [tvm] sunggg commented on a diff in pull request #12628: [MetaSchedule] Introduce `Union` and `OrderedUnion` in Database

[GitBox <gi...@apache.org>]

sunggg commented on code in PR #12628:
URL: https://github.com/apache/tvm/pull/12628#discussion_r960742815


##########
include/tvm/meta_schedule/database.h:
##########
@@ -357,6 +357,22 @@ class Database : public runtime::ObjectRef {
    */
   TVM_DLL static Database JSONDatabase(String path_workload, String path_tuning_record,
                                        bool allow_missing);
+  /*!
+   * \brief A database composed of multiple databases, allowing users to guide IR rewriting using
+   * combined knowledge of those databases. To each query, it returns the best record among all the
+   * databases given.
+   * \param databases The list of databases to be combined.
+   * \return The combined database.
+   */
+  TVM_DLL static Database UnionDatabase(Array<Database, void> databases);

Review Comment:
   Discussed offline with @junrushao and learned about more context behind the naming. 
   I'm okay with the naming, but would like to suggest more documentation with examples for better guidance.
   Some examples we could use:
   ```python
   ### Examples 
   Assumption: 
   * db1, d2 does not have tuning records for the target workload. 
   * each db3, db4, db5 have tuning records r3, r4, r5 for target workload respectively.  
   
   Case 1. `UnionDatabase`:  
   merged_db = ms.database.UnionDatabase(                    
       db1, # no record
       db2, # no record
       db3, # has r3
       db4  # has r4
   )
   # This return best one between r3 and r4
   merged_db.query_tuning_record(..., target_workload) 
   
   Case 2. `OrderedUnionDatabase`
   merged_db = ms.database.OrderedUnionDatabase(                    
       db1, # no record
       db2, # no record
       db3, # has r3
       db4  # has r4
   )
   # This return r3
   merged_db.query_tuning_record(..., target_workload) 
   
   Case 3. Mix-use scenario
   merged_db = ms.database.UnionDatabase(                    
       db1, # no record
       db2, # no record
       db3, # has r3
       ms.database.OrderedUnionDatabase( # returns r4
            db4,  # has r4
            db5,  # has r5
       )
   )
   # This return best one between r3 and r4
   merged_db.query_tuning_record(..., target_workload) 
   
   Case 4. Another mix-use scenario
   merged_db = ms.database.UnionDatabase(                    
       db1, # no record
       db2, # no record
       db3, # has r3
       ms.database.UnionDatabase( # returns best one between r4 and r5
            db4,  # has r4
            db5,  # has r5
       )
   )
   # This return best one among r3, r4 and r5
   merged_db.query_tuning_record(..., target_workload) 
   
   Case 5. Yet another mix-use scenario
   merged_db = ms.database.OrderedUnionDatabase(                    
       db1, # no record
       db2, # no record
       ms.database.UnionDatabase( # returns best one between r3 and r4
            db3, # has r3
            db4,  # has r4
       )
       db5,  # has r5
   )
   # This return best one between r3 and r4
   merged_db.query_tuning_record(..., target_workload) 
   ```



-- 
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: commits-unsubscribe@tvm.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org