Support limiting num sequences per category.

Summary:
Adds stratified sampling of sequences within categories applied after category / sequence filters but before the num sequence limit.
It respects the insertion order into the sequence_annots table, i.e. takes top N sequences within each category.

Reviewed By: bottler

Differential Revision: D46724002

fbshipit-source-id: 597cb2a795c3f3bc07f838fc51b4e95a4f981ad3

pytorch3d/implicitron/dataset/sql_dataset.py

@@ -89,6 +89,9 @@ class SqlIndexDataset(DatasetBase, ReplaceableBase):  # pyre-ignore
         pick_categories: Restrict the dataset to the given list of categories.
         pick_sequences: A Sequence of sequence names to restrict the dataset to.
         exclude_sequences: A Sequence of the names of the sequences to exclude.
+        limit_sequences_per_category_to: Limit the dataset to the first up to N
+            sequences within each category (applies after all other sequence filters
+            but before `limit_sequences_to`).
         limit_sequences_to: Limit the dataset to the first `limit_sequences_to`
             sequences (after other sequence filters have been applied but before
             frame-based filters).
@@ -115,6 +118,7 @@ class SqlIndexDataset(DatasetBase, ReplaceableBase):  # pyre-ignore
 
     pick_sequences: Tuple[str, ...] = ()
     exclude_sequences: Tuple[str, ...] = ()
+    limit_sequences_per_category_to: int = 0
     limit_sequences_to: int = 0
     limit_to: int = 0
     n_frames_per_sequence: int = -1
@@ -373,6 +377,7 @@ class SqlIndexDataset(DatasetBase, ReplaceableBase):  # pyre-ignore
             self.remove_empty_masks
             or self.limit_to > 0
             or self.limit_sequences_to > 0
+            or self.limit_sequences_per_category_to > 0
             or len(self.pick_sequences) > 0
             or len(self.exclude_sequences) > 0
             or len(self.pick_categories) > 0
@@ -380,20 +385,38 @@ class SqlIndexDataset(DatasetBase, ReplaceableBase):  # pyre-ignore
         )
 
     def _get_filtered_sequences_if_any(self) -> Optional[pd.Series]:
-        # maximum possible query: WHERE category IN 'self.pick_categories'
+        # maximum possible filter (if limit_sequences_per_category_to == 0):
+        # WHERE category IN 'self.pick_categories'
         # AND sequence_name IN 'self.pick_sequences'
         # AND sequence_name NOT IN 'self.exclude_sequences'
         # LIMIT 'self.limit_sequence_to'
 
-        stmt = sa.select(SqlSequenceAnnotation.sequence_name)
-
         where_conditions = [
             *self._get_category_filters(),
             *self._get_pick_filters(),
             *self._get_exclude_filters(),
         ]
-        if where_conditions:
+
-            stmt = stmt.where(*where_conditions)
+        def add_where(stmt):
+            return stmt.where(*where_conditions) if where_conditions else stmt
+
+        if self.limit_sequences_per_category_to <= 0:
+            stmt = add_where(sa.select(SqlSequenceAnnotation.sequence_name))
+        else:
+            subquery = sa.select(
+                SqlSequenceAnnotation.sequence_name,
+                sa.func.row_number()
+                .over(
+                    order_by=sa.text("ROWID"),  # NOTE: ROWID is SQLite-specific
+                    partition_by=SqlSequenceAnnotation.category,
+                )
+                .label("row_number"),
+            )
+
+            subquery = add_where(subquery).subquery()
+            stmt = sa.select(subquery.c.sequence_name).where(
+                subquery.c.row_number <= self.limit_sequences_per_category_to
+            )
 
         if self.limit_sequences_to > 0:
             logger.info(
@@ -402,7 +425,11 @@ class SqlIndexDataset(DatasetBase, ReplaceableBase):  # pyre-ignore
             # NOTE: ROWID is SQLite-specific
             stmt = stmt.order_by(sa.text("ROWID")).limit(self.limit_sequences_to)
 
-        if not where_conditions and self.limit_sequences_to <= 0:
+        if (
+            not where_conditions
+            and self.limit_sequences_to <= 0
+            and self.limit_sequences_per_category_to <= 0
+        ):
             # we will not need to filter by sequences
             return None
 

tests/implicitron/test_sql_dataset.py

@@ -222,6 +222,30 @@ class TestSqlDataset(unittest.TestCase):
         )
         self.assertEqual(len(dataset), 100)
 
+    def test_limit_sequence_per_category(self, num_sequences=2):
+        dataset = SqlIndexDataset(
+            sqlite_metadata_file=METADATA_FILE,
+            remove_empty_masks=False,
+            limit_sequences_per_category_to=num_sequences,
+            frame_data_builder_FrameDataBuilder_args=NO_BLOBS_KWARGS,
+        )
+
+        self.assertEqual(len(dataset), num_sequences * 10 * 2)
+        seq_names = list(dataset.sequence_names())
+        self.assertEqual(len(seq_names), num_sequences * 2)
+        # check that we respect the row order
+        for seq_name in seq_names:
+            self.assertLess(int(seq_name[-1]), num_sequences)
+
+        # test when the limit is not binding
+        dataset = SqlIndexDataset(
+            sqlite_metadata_file=METADATA_FILE,
+            remove_empty_masks=False,
+            limit_sequences_per_category_to=13,
+            frame_data_builder_FrameDataBuilder_args=NO_BLOBS_KWARGS,
+        )
+        self.assertEqual(len(dataset), 100)
+
     def test_filter_medley(self):
         dataset = SqlIndexDataset(
             sqlite_metadata_file=METADATA_FILE,