returnn 1.20241120.155202__tar.gz → 1.20241126.202240__tar.gz

{returnn-1.20241120.155202 → returnn-1.20241126.202240}/CODEOWNERS

@@ -5,14 +5,14 @@
 # Use absolute names when necessary or more helpful to identify.
 # Otherwise the basename is fine. Or other patterns.
 
-* @rwth-i6/returnn @albertz
-/returnn/tf	@albertz
-/returnn/theano	@albertz
+* @rwth-i6/returnn @albertz @NeoLegends
+/returnn/tf @albertz
+/returnn/theano @albertz
 
 better_exchook.py @albertz
 
 cached.py @patrick-wilken @JackTemaki
-hdf.py  @patrick-wilken @JackTemaki
+hdf.py @patrick-wilken @JackTemaki
 
 # Could be extended a lot more, but not sure if needed.
 # The global owners are also almost always the owners of more specific things.

{returnn-1.20241120.155202 → returnn-1.20241126.202240}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: returnn
-Version: 1.20241120.155202
+Version: 1.20241126.202240
 Summary: The RWTH extensible training framework for universal recurrent neural networks
 Home-page: https://github.com/rwth-i6/returnn/
 Author: Albert Zeyer

returnn-1.20241126.202240/_setup_info_generated.py

@@ -0,0 +1,2 @@
+version = '1.20241126.202240'
+long_version = '1.20241126.202240+git.88f3998'

{returnn-1.20241120.155202 → returnn-1.20241126.202240}/returnn/datasets/hdf.py

@@ -1075,7 +1075,17 @@ class SimpleHDFWriter:
     Note that we dump to a temp file first, and only at :func:`close` we move it over to the real destination.
     """
 
-    def __init__(self, filename, dim, labels=None, ndim=None, extra_type=None, swmr=False, extend_existing_file=False):
+    def __init__(
+        self,
+        filename,
+        dim,
+        labels=None,
+        ndim=None,
+        extra_type=None,
+        swmr=False,
+        extend_existing_file=False,
+        extra_labels=None,
+    ):
         """
         :param str filename: Create file, truncate if exists
         :param int|None dim:
@@ -1084,6 +1094,7 @@ class SimpleHDFWriter:
         :param dict[str,(int,int,str)]|None extra_type: key -> (dim,ndim,dtype)
         :param bool swmr: see https://docs.h5py.org/en/stable/swmr.html
         :param bool extend_existing_file: True also means we expect that it exists
+        :param dict[str,list[str]]|None extra_labels: key -> labels
         """
         from returnn.util.basic import hdf5_strings, unicode
         import tempfile
@@ -1147,7 +1158,7 @@ class SimpleHDFWriter:
         self._extra_num_time_steps = {}  # type: typing.Dict[str,int]  # key -> num-steps
         self._prepared_extra = set()
         if extra_type:
-            self._prepare_extra(extra_type)
+            self._prepare_extra(extra_type, extra_labels if extra_labels else {})
 
         if swmr:
             assert not self._file.swmr_mode  # this also checks whether the attribute exists (right version)
@@ -1160,7 +1171,7 @@ class SimpleHDFWriter:
             self._file.close()
             self._file = None
 
-    def _prepare_extra(self, extra_type):
+    def _prepare_extra(self, extra_type, extra_labels):
         """
         :param dict[str,(int,int,str)] extra_type: key -> (dim,ndim,dtype)
         :return: whether we added a new entry
@@ -1178,7 +1189,11 @@ class SimpleHDFWriter:
                 self._file.create_group("targets/data")
                 self._file.create_group("targets/size")
                 self._file.create_group("targets/labels")
-            hdf5_strings(self._file, "targets/labels/%s" % data_key, ["dummy-label"])
+            labels = ["dummy-label"]
+            if data_key in extra_labels:
+                labels = extra_labels[data_key]
+                assert len(labels) == dim
+            hdf5_strings(self._file, "targets/labels/%s" % data_key, labels)
             if ndim == 0:
                 ndim = 1  # we will automatically add a dummy-dim
             shape = [None] * ndim  # type: typing.List[typing.Optional[int]]
@@ -1261,7 +1276,7 @@ class SimpleHDFWriter:
             dtype = "string"
         else:
             dtype = raw_data.dtype.name
-        if self._prepare_extra({data_key: (dim, raw_data.ndim, dtype)}):
+        if self._prepare_extra({data_key: (dim, raw_data.ndim, dtype)}, {}):
             # We added it now. Maybe other extra data keys were added before. The data_key_idx is different now.
             # Thus, seq_lengths might have become invalid. Reinit them.
             assert seq_idx == 0 or self.extend_existing_file  # We can only do that in the beginning.

{returnn-1.20241120.155202 → returnn-1.20241126.202240}/returnn/datasets/postprocessing.py

@@ -241,7 +241,7 @@ class PostprocessingDataset(CachedDataset2):
 
         data_iter = self._iterate_dataset()
         if self._map_seq_stream is not None:
-            data_iter = self._map_seq_stream(data_iter, rng=self._rng, **util.get_fwd_compat_kwargs())
+            data_iter = self._map_seq_stream(data_iter, epoch=self.epoch, rng=self._rng, **util.get_fwd_compat_kwargs())
             assert isinstance(
                 data_iter, Iterator
             ), f"map_seq_stream must produce an {Iterator.__name__}, but produced {type(data_iter).__name__}"
@@ -263,7 +263,9 @@ class PostprocessingDataset(CachedDataset2):
             tensor_dict.data["seq_tag"].raw_tensor = str_to_numpy_array(self._dataset.get_tag(seq_index))
 
             if self._map_seq is not None:
-                tensor_dict = self._map_seq(tensor_dict, rng=self._rng, **util.get_fwd_compat_kwargs())
+                tensor_dict = self._map_seq(
+                    tensor_dict, epoch=self.epoch, seq_idx=seq_index, rng=self._rng, **util.get_fwd_compat_kwargs()
+                )
                 assert isinstance(
                     tensor_dict, TensorDict
                 ), f"map_seq must produce a {TensorDict.__name__}, but produced {type(tensor_dict).__name__}"

{returnn-1.20241120.155202 → returnn-1.20241126.202240}/returnn.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: returnn
-Version: 1.20241120.155202
+Version: 1.20241126.202240
 Summary: The RWTH extensible training framework for universal recurrent neural networks
 Home-page: https://github.com/rwth-i6/returnn/
 Author: Albert Zeyer

returnn-1.20241120.155202/_setup_info_generated.py

@@ -1,2 +0,0 @@
-version = '1.20241120.155202'
-long_version = '1.20241120.155202+git.2d96e42'