add MLP estimator, update Estimator generics

README.md

@@ -1,5 +1,6 @@
 # Overview
-Package summary goes here, ideally with a diagram
+Minimal framework for ML modeling, supporting advanced dataset operations and
+streamlined training workflows.
 
 # Install
 The `trainlib` package can be installed from PyPI:
@@ -85,7 +86,7 @@ pip install trainlib
   class SequenceDataset[I, **P](HomogenousDataset[int, I, I, P]):
       ...
   
-  class TupleDataset[I](SequenceDataset[tuple[I, ...], ??]):
+  class TupleDataset[I](SequenceDataset[tuple[I, ...], "?"]):
       ...
   ```
 

doc/conf.py

@@ -3,29 +3,64 @@
 # For the full list of built-in configuration values, see the documentation:
 # https://www.sphinx-doc.org/en/master/usage/configuration.html
 
+
+# -- Styling: type hints ------------------------------------------------------
+# There are several possible style combinations for rendering types, none of
+# which are optimal in my view. The main switches are:
+# 
+# - Parameter type hints in the signature vs in the separate parameter list
+# - Show type hints as plaintext vs rendered HTML elements
+#
+# The `sphinx_autodoc_typehints` extension enables more context-aware
+# rendering, but it's often way too explicit (e.g., unwrapping type variables)
+# and makes things difficult to read. It does, however, allow for automatic
+# inclusion of default values, which is nice.
+#
+# I'd like type hints to be rendered in an inline code element, but that
+# doesn't happen by default in either case unless you render them in the
+# signature. This is sloppy, however, often just a jumbled mess or parameter
+# names and types. The current preferred option is to just use the native
+# `autodoc` settings for rendering type hints, leaving them out of the
+# signature (for easy heading readability). Type hints in the parameter list
+# are also as short as possible, not rendered crazily (by default this is in
+# italics; not my favorite but it's what we have). No
+# `sphinx_autodoc_typehints` needed at this point; you can toggle it if you
+# want automatic default values or different formatting for type hints.
+
+
 # -- Project information ------------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#project-information
 
-project = "<package-name>"
+project = "trainlib"
-copyright = "2025, Sam Griesemer"
+copyright = "2026, Sam Griesemer"
 author = "Sam Griesemer"
 
+
 # -- General configuration ----------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#general-configuration
 
 extensions = [
     "sphinx.ext.autodoc",
+
     # enables a directive to be specified manually that gathers module/object
     # summary details in a table
     "sphinx.ext.autosummary",
+
     # allow viewing source in the HTML pages
     "sphinx.ext.viewcode",
+
     # only really applies to manual docs; docstrings still need RST-like
     "myst_parser",
+
     # enables Google-style docstring formats
     "sphinx.ext.napoleon",
-    # external extension that allows arg types to be inferred by type hints
+
-    "sphinx_autodoc_typehints",
+    # external extension that allows arg types to be inferred by type hints;
+    # without this, type hints show up inside method signatures as plaintext,
+    # but when enabled they are pulled into the parameter/description block and
+    # rendered as native nested markup. What's best for a given package may
+    # vary.
+    # "sphinx_autodoc_typehints",
 ]
 autosummary_generate = True
 autosummary_imported_members = True
@@ -39,11 +74,39 @@ templates_path = ["_templates"]
 exclude_patterns = ["_build", "Thumbs.db", ".DS_Store"]
 
 
+# -- Options for autodoc ------------------------------------------------------
+# class signatures show up only in __init__ rather than at the class header;
+# generally cleaner, avoids redundancy
+autodoc_class_signature = "separated"
+
+# if `sphinx_autodoc_typehints` extension is enabled, this is redundant: type
+# hints are rendered natively and already show up in the parameter block. If
+# it's disabled, this setting will do the same job of moving the types to the
+# parameter block, but it renders them in plaintext (with links to in-package
+# type refs).
+autodoc_typehints = "description"  # "signature"
+autodoc_typehints_format = "short"
+autodoc_preserve_defaults = True
+autodoc_use_type_comments = False
+python_use_unqualified_type_names = True
+
+# push parameters to their own lines in the signature block
+# python_maximum_signature_line_length = 60
+
+
+# -- Options for autodoc_typehints --------------------------------------------
+# always_use_bars_union = True  # always on for Python 3.14+
+# typehints_defaults = "braces-after"  # render defaults in param block
+# typehints_use_signature = False  # False is default; enable if wanted in sig
+# always_document_param_types = True  # show types even when not in docstring
+
+
 # -- Options for HTML output --------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#options-for-html-output
 
-html_theme = "furo"
+html_theme = "furo"  # "pydata_sphinx_theme"
 html_static_path = ["_static"]
+
 # html_sidebars = {
 #    '**': ['/modules.html'],
 # }

doc/index.md

@@ -1,29 +1,37 @@
-# `<project-name>` package docs 
+# `trainlib` package docs 
+
 {ref}`genindex`
 {ref}`modindex`
-{ref}`search`
 
 ```{eval-rst}
 .. autosummary::
    :nosignatures:
+   :recursive:
+   :caption: Modules
 
-    # list modules here for quick links
+    trainlib.dataset
+    trainlib.domain
+    trainlib.estimator
+    trainlib.trainer
+    trainlib.transform
 ```
 
 ```{toctree}
 :maxdepth: 3
 :caption: Autoref
+:hidden:
 
-_autoref/<project-name>.rst
+_autoref/trainlib.rst
 ```
 
 ```{toctree}
 :maxdepth: 3
 :caption: Contents
+:hidden:
 
 reference/documentation/index
-reference/site/index
 ```
 
 ```{include} ../README.md
+:heading-offset: 1
 ```

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "trainlib"
-version = "0.1.1"
+version = "0.1.2"
 description = "Minimal framework for ML modeling. Supports advanced dataset operations and streamlined training."
 requires-python = ">=3.13"
 authors = [
@@ -24,11 +24,11 @@ classifiers = [
     "Intended Audience :: End Users/Desktop",
 ]
 dependencies = [
+    "torch",
     "colorama>=0.4.6",
     "matplotlib>=3.10.8",
     "numpy>=2.4.1",
     "tensorboard>=2.20.0",
-    "torch>=2.5.1",
     "tqdm>=4.67.1",
 ]
 
@@ -41,6 +41,7 @@ dev = [
 ]
 doc = [
     "furo",
+    # "pydata-sphinx-theme",
     "myst-parser",
     "sphinx",
     "sphinx-togglebutton",
@@ -82,3 +83,11 @@ force-sort-within-sections = false
 quote-style = "double"
 indent-style = "space"
 docstring-code-format = true
+
+[tool.uv.sources]
+torch = { index = "pytorch" }
+
+[[tool.uv.index]]
+name = "pytorch"
+url = "https://download.pytorch.org/whl/cu128"
+explicit = true

trainlib/dataset.py

@@ -1,128 +1,62 @@
 """
-.. admonition:: Marginalizing out the modality layer
+Domain-generic dataset base with attribute-based splitting and balancing
 
-    With ``domain`` being an instance variable, one possible interpretation of
+**Marginalizing out the modality layer**
-    the object structures here is that one could completely abstract away
-    the domain model, defining only item structures and processing data. You
-    could have a single dataset definition for a particular concrete dataset,
-    and so long as we're talking about the same items, it can be instantiated
-    using *any domain*. You wouldn't need specific subclasses for disk or
-    network or in-memory; you can tell it directly at runtime.
 
-    That's an eventually possibility, anyway. As it stands, however, this is
+With ``domain`` being an instance variable, one possible interpretation of
-    effectively impossible:
+the object structures here is that one could completely abstract away
+the domain model, defining only item structures and processing data. You
+could have a single dataset definition for a particular concrete dataset,
+and so long as we're talking about the same items, it can be instantiated
+using *any domain*. You wouldn't need specific subclasses for disk or
+network or in-memory structures; you can tell it directly at runtime.
 
-    You can't easily abstract the batch -> item splitting process, i.e.,
+That's an eventually possibility, anyway. As it stands, however, this is
-    ``_process_batch_data()``. A list-based version of the dataset you're
+effectively impossible:
-    trying to define might have an individual item tuple at every index,
-    whereas a disk-based version might have tuples batched across a few files.
-    This can't reliably be inferred, nor can it be pushed to the
-    ``Domain``-level without needing equal levels of specialization (you'd just
-    end up needing the exact same structural distinctions in the ``Domain``
-    hierarchy). So *somewhere* you need a batch splitting implementation that
-    is both item structure-dependent *and* domain-dependent...the question is
-    how dynamic you're willing to be about where it comes from. Right now, we
-    require this actually be defined in the ``_process_batch_data()`` method,
-    meaning you'll need a specific ``Dataset`` class for each domain you want
-    to support (e.g., ``MNISTDisk``, ``MNISTList``, ``MNISTNetwork``, etc), or
-    at least for each domain where "interpreting" a batch could possibly
-    differ. This is a case where the interface is all that enforces a
-    distinction: if you've got two domains that can be counted on to yield
-    batches in the exact same way and can use the same processing, then you
-    could feasibly provide ``Domain`` objects from either at runtime and have
-    no issues. We're "structurally blind" to any differentiation beyond the URI
-    and resource types by design, so two different domain implementations with
-    the same type signature ``Domain[U, R]`` should be expected to work fine at
-    runtime (again, so long as they don't also need different batch
-    processing), but that's not affording us much flexibility, i.e., most of
-    the time we'll still be defining new dataset classes for each domain.
 
-    I initially flagged this as feasible, however, because one could imagine
+You can't easily abstract the batch-to-item splitting process, i.e.,
-    accepting a batch processing method upon instantiation rather than
+``_process_batch_data()``. A list-based version of the dataset you're trying to
-    structurally bolting it into the ``Dataset`` definition. This would require
+define might have an individual item tuple at every index, whereas a disk-based
-    knowledge of the item structure ``I`` as well as the ``Domain[U, R]``, so
+version might have tuples batched across a few files. This can't reliably be
-    such a function will always have to be (I, U, R)-dependent. It nevertheless
+inferred, nor can it be pushed to the ``Domain``-level without needing equal
-    would take out some of the pain of having to define new dataset classes;
+levels of specialization (you'd just end up needing the exact same structural
-    instead, you'd just need to define the batch processing method. I see this
+distinctions in the ``Domain`` hierarchy). So *somewhere* you need a batch
-    as a worse alternative to just defining *inside* a safe context like a new
+splitting implementation that is both item structure-dependent *and*
-    dataset class: you know the types you have to respect, and you stick that
+domain-dependent...the question is how dynamic you're willing to be about where
-    method exactly in a context where it's understood. Freeing this up doesn't
+it comes from. Right now, we require this actually be defined in the
-    lighten the burden of processing logic, it just changes *when* it has to be
+``_process_batch_data()`` method, meaning you'll need a specific ``Dataset``
-    provided, and that's not worth much (to me) in this case given the bump in
+class for each domain you want to support (e.g., ``MNISTDisk``, ``MNISTList``,
-    complexity. (Taking this to the extreme: you could supply *all* of an
+``MNISTNetwork``, etc), or at least for each domain where "interpreting" a
-    object's methods "dynamically" and glue them together at runtime so long as
+batch could possibly differ. This is a case where the interface is all that
-    they all played nice. But wherever you were "laying them out" beforehand is
+enforces a distinction: if you've got two domains that can be counted on to
-    exactly the job of a class to begin with, so you don't end up with anything
+yield batches in the exact same way and can use the same processing, then you
-    more dynamic. All we're really discussing here is pushing around
+could feasibly provide ``Domain`` objects from either at runtime and have no
-    unavoidable complexity inside and outside of the "class walls," and in the
+issues. We're "structurally blind" to any differentiation beyond the URI and
-    particular case of ``_process_batch_data()``, it feels much better when
+resource types by design, so two different domain implementations with the same
-    it's on the inside.)
+type signature ``Domain[U, R]`` should be expected to work fine at runtime
+(again, so long as they don't also need different batch processing), but that's
+not affording us much flexibility, i.e., most of the time we'll still be
+defining new dataset classes for each domain.
 
-.. admonition:: Holding area
+I initially flagged this as feasible, however, because one could imagine
-
+accepting a batch processing method upon instantiation rather than structurally
-    .. code-block:: python
+bolting it into the ``Dataset`` definition. This would require knowledge of the
-
+item structure ``I`` as well as the ``Domain[U, R]``, so such a function will
-        @abstractmethod
+always have to be ``(I, U, R)``-dependent. It nevertheless would take out some
-        def _get_uri_groups(self) -> Iterable[tuple[U, ...]]:
+of the pain of having to define new dataset classes; instead, you'd just need
-            Get URI groups for each batch.
+to define the batch processing method. I see this as a worse alternative to
-
+just defining *inside* a safe context like a new dataset class: you know the
-            If there's more than one URI per batch (e.g., a data file and a
+types you have to respect, and you stick that method exactly in a context where
-            metadata file), zip the URIs such that we have a tuple of URIs per
+it's understood. Freeing this up doesn't lighten the burden of processing
-            batch.
+logic, it just changes *when* it has to be provided, and that's not worth much
-
+(to me) in this case given the bump in complexity. (Taking this to the extreme:
-            Note that this effectively defines the index style over batches in
+you could supply *all* of an object's methods "dynamically" and glue them
-            the attached domain. We get an ``int -> tuple[U, ...]`` map that
+together at runtime so long as they all played nice. But wherever you were
-            turns batch indices into URIs that can be read under the domain.
+"laying them out" beforehand is exactly the job of a class to begin with, so
-            ``get_batch()`` turns an integer index into its corresponding
+you don't end up with anything more dynamic. All we're really discussing here
-            ``tuple[U, ...]``, reading the resources with ``_read_resources()``
+is pushing around unavoidable complexity inside and outside of the "class
-            in the tuple, treating them as providers of batched data.
+walls," and in the particular case of ``_process_batch_data()``, it feels much
-            ``_read_resources()`` passes through to the attached domain logic,
+better when it's on the inside.)
-            which, although common, need not supply an explicit iterable of
-            batch items: we just access items with ``__getitem__()`` and may
-            ask for ``__len__``. So the returned URI group collection (this
-            method) does need to be iterable to measure the number of batches,
-            but the batch objects that are ultimately produced by these URI
-            groups need not be iterables themselves.
-
-            raise NotImplementedError
-
-        def _read_resources(
-            self,
-            uri_group: tuple[U, ...],
-            batch_index: int
-        ) -> tuple[R, ...]:
-            Read batch files at the provided paths.
-
-            This method should operate on a single tuple from the list of batch
-            tuples returned by the ``_get_uri_groups()`` method. That is, it
-            reads all of the resources for a single batch and returns a tuple
-            of the same size with their contents.
-
-            Note: the dependence on a batch index is mostly here to make
-            multi-dataset composition easier later. In-dataset, you don't need
-            to know the batch index to to simply process URIs, but across
-            datasets you need it to find out the origin of the batch (and
-            process those URIs accordingly).
-
-            return tuple(self.domain.read(uri) for uri in uri_group)
-
-    .. code-block:: python
-
-        # pulling the type variable out of the inline generic b/c `ty` has
-        # trouble understanding bound type variables in subclasses
-        # (specifically with Self@)
-
-        T = TypeVar("T", bound=NamedTuple)
-
-        class NamedTupleDataset[I](Dataset):
-            def __init__(self, data_list: list[I]) -> None:
-                self.data_list = data_list
-
-            def __len__(self) -> int:
-                return len(self.data_list)
-
-            def __getitem__(self, index: int) -> I:
-                return self.data_list[index]
 """
 
 import math
@@ -161,35 +95,69 @@ class BatchedDataset[U, R, I](Dataset):
 
     The class is generic over a URI type ``U``, a resource type ``R`` (both of
     which are used to concretize a domain ``Domain[U, R]``), and an item type
-    ``T`` (which has a ``tuple`` upper bound).
+    ``I``.
 
+    **Batch and item processing flow**
 
-    .. admonition:: Pipeline overview
+    .. code-block:: text
 
-        .. code-block:: python
+        Domain -> [U] :: self._batch_uris = list(domain)
 
-            Domain -> [U] (get _batch_uris)
+            Grab all URIs from Domain iterators. This is made concrete early to
-            U -> R (domain access ; Rs provide batches)
+            allow for Dataset sizing, and we need a Sequence representation to
-            R -> [I] (cache here ; _process_batch_data to use load_transform)
+            map integer batch indices into Domains, i.e., when getting the
-            [I] -> I (human item obj ; _get_item)
+            corresponding URI:
-            I -> **P (final packed item ; __getitem__ to use transform)
 
-        Note^1: as far as positioning, this class is meant to play nice with
+            batch_uri = self._batch_uris[batch_index]
-        PyTorch DataLoaders, hence the inheritance from ``torch.Dataset``. The
-        value add for this over the ``torch.Dataset`` base is almost entirely
-        in the logic it implements to map out of *batched resources* that are
-        holding data, and flattening it out into typical dataset items. There
-        are also some QoL items when it comes to splitting and balancing
-        samples. 
 
-        Note^2: even though ``Domains`` implement iterators over their URIs,
+            We let Domains implement iterators over their URIs, but explicitly
-        this doesn't imply a ``BatchedDataset`` is iterable. This just means we
+            exhaust when initializing Datasets.
-        can walk over the resources that provide data, but we don't necessarily
+
+        U -> R        :: batch_data = self.domain[batch_uri]
+
+            Retrieve resource from domain. Resources are viewed as batched
+            data, even if only wrapping single items (happens in trivial
+            settings).
+
+        R -> [I]      :: self._process_batch_data(batch_data, batch_index)
+
+            Possibly domain-specific batch processing of resource data into
+            explicit Sequence-like structures of items, each of which is
+            subject to the provided pre_transform. Processed batches at this
+            stage are cached (if enabled).
+
+        [I] -> I      :: self.get_batch(batch_index)[index_in_batch]
+
+            Select individual items from batches in _get_item. At this stage,
+            items are in intermediate states and pulled from the cached
+            batches.
+
+        I -> I        :: self._process_item_data(item_data, index)
+            
+            Produce final items with __getitem__, getting intermediate items
+            via _get_item and applying the provided post_transform.
+
+    .. note::
+
+        As far as positioning, this class is meant to play nice with PyTorch
+        DataLoaders, hence the inheritance from ``torch.Dataset``. The value
+        add for this over the ``torch.Dataset`` base is almost entirely in the
+        logic it implements to map out of *batched resources* that are holding
+        data, and flattening it out into typical dataset items. There are also
+        some QoL features when it comes to splitting and balancing samples.
+
+    .. note::
+
+        Even though ``Domains`` implement iterators over their URIs, this
+        doesn't imply a ``BatchedDataset`` is iterable. This just means we can
+        walk over the resources that provide data, but we don't necessarily
         presuppose an ordered walk over samples within batches. Point being:
         ``torch.Dataset``, not ``torch.IterableDataset``, is the appropriate
         superclass, even when we're working around iterable ``Domains``.
 
-        Note^3: transforms are expected to operate on ``I``-items and produce
+    .. note::
+
+        Transforms are expected to operate on ``I``-items and produce
         ``I``-items. They shouldn't be the "introducers" of ``I`` types from
         some other intermediate representation, nor should they map from ``I``
         to something else. Point being: the dataset definition should be able
@@ -211,6 +179,7 @@ class BatchedDataset[U, R, I](Dataset):
     ) -> None:
         """
         Parameters:
+            domain: ``Domain`` object providing access to batched data
             pre_transform: transform to apply over items during loading (in
                 ``_process_batch_data()``), i.e., *before* going into
                 persistent storage
@@ -220,6 +189,7 @@ class BatchedDataset[U, R, I](Dataset):
             batch_cache_limit: the max number of max batches to cache at any
                 one time
             preload: whether to load all data into memory during instantiation
+            num_workers: number of workers to use when preloading data
         """
 
         self.domain = domain
@@ -259,6 +229,9 @@ class BatchedDataset[U, R, I](Dataset):
         The behavior of this method can vary depending on what we know about
         batch sizes, and should therefore be implemented by inheriting classes.
 
+        Parameters:
+            item_index: index of item
+
         Returns:
             batch_index: int
             index_in_batch: int
@@ -302,6 +275,10 @@ class BatchedDataset[U, R, I](Dataset):
         place to use a provided ``post_transform``; items are pulled from the
         cache (if enabled) and processed before being returned as the final
         tuple outputs (so this processing is not persistent).
+
+        Parameters:
+            item_data: item data
+            item_index: index of item
         """
 
         raise NotImplementedError
@@ -317,6 +294,9 @@ class BatchedDataset[U, R, I](Dataset):
 
         Note that return values from `__getitem__()` are "cleaned up" versions
         of this representation, with minimal info needed for training.
+
+        Parameters:
+            item_index: index of item
         """
 
         if item_index >= len(self):
@@ -351,10 +331,13 @@ class BatchedDataset[U, R, I](Dataset):
         they're always connected, and nothing would notice if you waited
         between steps. The only way this could matter is if you split the
         resource reading and batch processing steps across methods, but when it
-        actually comes to accessing/caching the batch, you'd have to expand
+        actually comes to accessing/caching the batch, you'd have to expand any
-        any delayed reads here. There's no way around needing to see all batch
+        delayed reads here. There's no way around needing to see all batch data
-        data at once here, and we don't want to make that ambiguous: ``list``
+        at once here, and we don't want to make that ambiguous: ``list`` output
-        output type it is.
+        type it is.
+
+        Parameters:
+            batch_index: index of batch
         """
 
         logger.debug("Batch cache miss, reading from root...")
@@ -374,6 +357,9 @@ class BatchedDataset[U, R, I](Dataset):
         Can be useful when dynamically pulling data (as it's requested) isn't
         desired. Requires that `cache_sample_limit=None`, i.e., the cache won't
         continually remove previous batches as they're loaded.
+
+        Parameters:
+            num_workers: number of parallel workers to use for data loading
         """
 
         assert self.batch_cache_limit is None, "Preloading under cache limit"
@@ -406,36 +392,46 @@ class BatchedDataset[U, R, I](Dataset):
         """
         Split dataset into fractional pieces by data attribute.
 
-        If `by_attr` is None, recovers typical fractional splitting of dataset
+        If ``by_attr`` is None, recovers typical fractional splitting of
-        items, partitioning by size. Using None anywhere will index each item
+        dataset items, partitioning by size. Using None anywhere will index
-        into its own bucket, i.e., by its index. For instance,
+        each item into its own bucket, i.e., by its index. For instance:
 
-        - by_attr=["color"] -> {("red", 1), ("red", 2)},
+        - Splits on the attribute such that each subset contains entire strata
+          of the attribute. "Homogeneity within clusters:"
+
+          .. code-block::
+
+              by_attr=["color"] -> {("red", 1), ("red", 2)},
                                    {("blue", 1), ("blue", 2)}
 
-          Splits on the attribute such that each subset contains entire strata
+        - Stratifies by attribute and then splits "by index" within, uniformly
-          of the attribute. "Homogeneity within clusters"
-
-        - `by_attr=["color", None]` -> {("red", 1), ("blue", 1)},
-                                       {("red", 2), ("blue", 2)}
-
-          Stratifies by attribute and then splits "by index" within, uniformly
           grabbing samples across strata to form new clusters. "Homogeneity
           across clusters"
 
+          .. code-block::
+
+              by_attr=["color", None] -> {("red", 1), ("blue", 1)},
+                                         {("red", 2), ("blue", 2)}
+
         Note that the final list of Subsets returned are built from shallow
-        copies of the underlying dataset (i.e., `self`) to allow manual
+        copies of the underlying dataset (i.e., ``self``) to allow manual
         intervention with dataset attributes (e.g., setting the splits to have
-        different `transform`s). This is subject to possibly unexpected
+        different ``transforms``). This is subject to possibly unexpected
         behavior if re-caching data or you need a true copy of all data in
         memory, but should otherwise leave most interactions unchanged.
 
         Parameters:
+            frac: split fractions for datasets
+            dataset: dataset to split, defaults to ``self``. Facilitates
+                recursive splitting when multi-attribute splits are needed.
+            by_attr: attribute or attributes to use when grouping strata for
+                dataset splits. Defaults to ``None``, which will use item
+                indices.
             shuffle_strata: shuffle the strata order before split is drawn. We
-                parameterize this because a dataloader-level shuffle operation
+                parameterize this because a Dataloader-level shuffle operation
                 will only change the order of the indices in the resulting
-                splits; only a shuffle of items inside the strata can change
+                splits; only a shuffle of the strata order can change the
-                the actual content of the splits themselves.
+                actual content of the splits themselves. 
         """
 
         if by_attr == []:
@@ -544,6 +540,32 @@ class BatchedDataset[U, R, I](Dataset):
         split_max_sizes: list[int] | None = None,
         shuffle_strata: bool = True,
     ) -> None:
+        """
+        Balance the distribution of provided attributes over dataset items.
+
+        This method sets the indices over the dataset according to the result
+        of the rebalancing. The indices are produced by the recursive
+        ``_balance()`` method, which is necessarily separate due to the need
+        for a contained recursive approach that doesn't change the underlying
+        dataset during execution.
+
+        Parameters:
+            dataset: dataset to split, defaults to ``self``. Facilitates
+                recursive splitting when multi-attribute splits are needed.
+            by_attr: attribute or attributes to use when grouping strata for
+                dataset splits. Defaults to ``None``, which will use item
+                indices.
+            split_min_sizes: minimum allowed sizes of splits. Must have the
+                same length as ``by_attr``. 
+            split_max_sizes: maximum allowed sizes of splits. Must have the
+                same length as ``by_attr``. 
+            shuffle_strata: shuffle the strata order before split is drawn. We
+                parameterize this because a Dataloader-level shuffle operation
+                will only change the order of the indices in the resulting
+                splits; only a shuffle of the strata order can change the
+                actual content of the splits themselves. 
+        """
+
         self.indices = self._balance(
             dataset,
             by_attr,
@@ -561,9 +583,29 @@ class BatchedDataset[U, R, I](Dataset):
         shuffle_strata: bool = True,
     ) -> list[int]:
         """
-        Note: behavior is a little odd for nested behavior; not exactly
+        Recursive balancing of items by attribute.
-        perfectly uniform throughout.  This is a little difficult: you can't
+
-        exactly know ahead of time the size of the subgroups across splits
+        .. note::
+
+            Behavior is a little odd for nested behavior; not exactly perfectly
+            uniform throughout. This is a little difficult: you can't exactly
+            know ahead of time the size of the subgroups across splits
+
+        Parameters:
+            dataset: dataset to split, defaults to ``self``. Facilitates
+                recursive splitting when multi-attribute splits are needed.
+            by_attr: attribute or attributes to use when grouping strata for
+                dataset splits. Defaults to ``None``, which will use item
+                indices.
+            split_min_sizes: minimum allowed sizes of splits. Must have the
+                same length as ``by_attr``. 
+            split_max_sizes: maximum allowed sizes of splits. Must have the
+                same length as ``by_attr``. 
+            shuffle_strata: shuffle the strata order before split is drawn. We
+                parameterize this because a Dataloader-level shuffle operation
+                will only change the order of the indices in the resulting
+                splits; only a shuffle of the strata order can change the
+                actual content of the splits themselves. 
         """
 
         if by_attr == []:
@@ -653,6 +695,9 @@ class BatchedDataset[U, R, I](Dataset):
         dataset. The underlying data remain the same, but when indices get set,
         you're effectively applying a mask over any existing indices, always
         operating *relative* to the existing mask.
+
+        Parameters:
+            indices: list of indices to set
         """
 
         # manually set new size
@@ -680,6 +725,13 @@ class BatchedDataset[U, R, I](Dataset):
         return self._dataset_len
 
     def __getitem__(self, index: int) -> I:
+        """
+        Get the dataset item at the specified index.
+
+        Parameters:
+            index: index of item to retrieve
+        """
+
         item_data = self._get_item(index)
         index = self.indices[index]
 
@@ -701,9 +753,10 @@ class CompositeBatchedDataset[U, R, I](BatchedDataset[U, R, I]):
     """
     Dataset class for wrapping individual datasets.
 
-    Note: because this remains a valid ``BatchedDataset``, we re-thread the
+    .. note::
-    generic type variables through the set of composed datasets. That is, they
+        Because this remains a valid ``BatchedDataset``, we re-thread the
-    must have a common domain type ``Domain[U, R]``.
+        generic type variables through the set of composed datasets. That is,
+        they must have a common domain type ``Domain[U, R]``.
     """
 
     def __init__(
@@ -888,7 +941,7 @@ class HomogenousDataset[U, R, I](BatchedDataset[U, R, I]):
 
 class HeterogenousDataset[U, R, I](BatchedDataset[U, R, I]):
     """
-    Batched dataset where batches have arbitrary size.
+    Batched dataset where batches may have arbitrary size.
 
     Methods left for inheriting classes:
 

trainlib/datasets/disk.py

@@ -12,7 +12,7 @@ class DiskDataset[T: NamedTuple](HomogenousDataset[Path, bytes, T]):
     """
     The following line is to satisfy the type checker, which
 
-    1. Can't recognize an appropriately re-typed constructor arg like
+    1. Can't recognize an appropriately re-typed constructor arg like::
 
            def __init__(
                self,
@@ -20,7 +20,8 @@ class DiskDataset[T: NamedTuple](HomogenousDataset[Path, bytes, T]):
                ...
            ): ...
 
-       This *does* match the parent generic for the U=Path, R=bytes context
+       This *does* match the parent generic for the ``U=Path``, ``R=bytes``
+       context::
 
            def __init__(
                self,
@@ -32,19 +33,17 @@ class DiskDataset[T: NamedTuple](HomogenousDataset[Path, bytes, T]):
 
     2. "Lifted" type variables out of generics can't be used as upper bounds,
        at least not without throwing type checker warnings (thanks to PEP695).
-       So I'm not allowed to have
+       So I'm not allowed to have::
 
-       ```
            class BatchedDataset[U, R, D: Domain[U, R]]:
                ...
-       ```
 
-       which could bring appropriately dynamic typing for ``Domain``s, but is
+       which could bring appropriately dynamic typing for ``Domains``, but is
        not a sufficiently concrete upper bound.
 
     So: we settle for a class-level type declaration, which despite not being
     technically appropriately scoped, it's not harming anything and satisfies
-    ``ty`` type checks downstream (e.g., when we access ``DiskDomain.root``.
+    ``ty`` type checks downstream (e.g., when we access ``DiskDomain.root``).
     """
 
     domain: DiskDomain

trainlib/datasets/memory.py

@@ -80,10 +80,10 @@ from trainlib.domain import SequenceDomain
 from trainlib.dataset import TupleDataset, DatasetKwargs
 
 
-class SlidingWindowDataset[T: Tensor](TupleDataset[T]):
+class SlidingWindowDataset(TupleDataset[Tensor]):
     def __init__(
         self,
-        domain: SequenceDomain[tuple[T, ...]],
+        domain: SequenceDomain[tuple[Tensor, ...]],
         lookback: int,
         offset: int = 0,
         lookahead: int = 1,
@@ -99,9 +99,9 @@ class SlidingWindowDataset[T: Tensor](TupleDataset[T]):
 
     def _process_batch_data(
         self,
-        batch_data: tuple[T, ...],
+        batch_data: tuple[Tensor, ...],
         batch_index: int,
-    ) -> list[tuple[T, ...]]:
+    ) -> list[tuple[Tensor, ...]]:
         """
         Backward pads first sequence over (lookback-1) length, and steps the
         remaining items forward by the lookahead.

trainlib/domain.py

@@ -1,24 +1,5 @@
 """
-Defines a knowledge domain. Wraps a Dataset / Simulator / Knowledge
+Generic URI-resource mapping structure
-
-Downstream exploration might include
-
-- Calibrating Simulator / Knowledge with a Dataset
-- Amending Dataset with Simulator / Knowledge
-- Positioning Knowledge within Simulator context
-  * Where to replace Simulator subsystem with Knowledge?
-
-Other variations:
-
-- Multi-fidelity simulators
-- Multi-scale models
-- Multi-system
-- Incomplete knowledge / divergence among sources
-
-Questions:
-
-- Should Simulator / Knowledge be unified as one (e.g., "Expert")
-
 """
 
 from collections.abc import Mapping, Iterator, Sequence
@@ -83,3 +64,14 @@ class SequenceDomain[R](Domain[int, R]):
 
     def __len__(self) -> int:
         return len(self.sequence)
+
+
+class TupleDomain[T](SequenceDomain[tuple[T, ...]]):
+    """
+    Domain for homogenous tuples of the same type.
+
+    This class header exists primarily as typed alias that aligns with
+    TupleDataset.
+    """
+
+    ...

trainlib/domains/functional.py

@@ -8,10 +8,16 @@ class SimulatorDomain[P, R](Domain[int, R]):
     Base simulator domain, generic to arbitrary callables.
 
     Note: we don't store simulation results here; that's left to a downstream
-    object, like a `BatchedDataset`, to cache if needed. We also don't subclass
+    object, like a ``BatchedDataset``, to cache if needed. We also don't
-    `SequenceDataset` because the item getter type doesn't align: we accept an
+    subclass ``SequenceDataset`` because the item getter type doesn't align: we
-    `int` in the parameter list, but don't return the items directly from that
+    accept an ``int`` in the parameter list, but don't return the items
-    collection (we transform them first).
+    directly from that collection (we transform them first).
+
+    Note: it's interesting to consider the idea of having parameters directly
+    act as URIs. There is, however, no obvious way to iterate over allowed
+    parameters (without additional components, like a prior or some other
+    generator), so we leave that outside the class scope and simply operate
+    over of a provided parameter sequence.
     """
 
     def __init__(

trainlib/estimator.py

@@ -1,4 +1,6 @@
 """
+Base class for trainable models
+
 Development note
 
 I'd rather lay out bare args and kwargs in the estimator methods, but the

trainlib/estimators/mlp.py

@@ -0,0 +1,149 @@
+import logging
+from typing import Unpack, NotRequired
+from collections.abc import Callable, Generator
+
+import torch
+import torch.nn.functional as F
+from torch import nn, Tensor
+from torch.optim import Optimizer
+from torch.utils.tensorboard import SummaryWriter
+
+from trainlib.estimator import Estimator, EstimatorKwargs
+from trainlib.utils.type import OptimizerKwargs
+from trainlib.utils.module import get_grad_norm
+from trainlib.estimators.tdnn import TDNNLayer
+
+logger: logging.Logger = logging.getLogger(__name__)
+
+
+class MLPKwargs(EstimatorKwargs):
+    inputs: Tensor
+    labels: NotRequired[Tensor]
+
+
+class MLP[K: MLPKwargs](Estimator[K]):
+    """
+    Base MLP architecture.
+    """
+
+    def __init__(
+        self,
+        input_dim: int,
+        output_dim: int,
+        hidden_dims: list[int] | None = None,
+        norm_layer: Callable[..., nn.Module] | None = None,
+        activation_fn: nn.Module | None = None,
+        inplace: bool = False,
+        bias: bool = True,
+        dropout: float = 0.0,
+        verbose: bool = True,
+    ) -> None:
+        """
+        Parameters:
+            input_dim: dimensionality of the input
+            output_dim: dimensionality of the output
+            hidden_dims: dimensionalities of hidden layers
+        """
+
+        super().__init__()
+
+        self.input_dim = input_dim
+        self.output_dim = output_dim
+        self.hidden_dims = hidden_dims or []
+        self.norm_layer = norm_layer
+        self.activation_fn = activation_fn or nn.ReLU
+
+        # self._layers: nn.ModuleList = nn.ModuleList()
+        self._layers = []
+
+        layer_in_dim = input_dim
+        for layer_out_dim in self.hidden_dims:
+            hidden_layer = nn.Linear(layer_in_dim, layer_out_dim, bias=bias)
+            self._layers.append(hidden_layer)
+
+            if norm_layer is not None:
+                self._layers.append(norm_layer(layer_out_dim))
+
+            self._layers.append(self.activation_fn(inplace=inplace))
+            self._layers.append(nn.Dropout(dropout, inplace=inplace))
+
+            layer_in_dim = layer_out_dim
+
+        self._layers.append(nn.Linear(layer_in_dim, self.output_dim))
+        self._net = nn.Sequential(*self._layers)
+
+        if verbose:
+            self.log_arch()
+
+    def _clamp_rand(self, x: Tensor) -> Tensor:
+        return torch.clamp(
+            x + (1.0 / 127.0) * (torch.rand_like(x) - 0.5),
+            min=-1.0,
+            max=1.0,
+        )
+
+    def forward(self, **kwargs: Unpack[K]) -> tuple[Tensor, ...]:
+        inputs = kwargs["inputs"]
+        x = self._net(inputs)
+
+        return (x,)
+
+    def loss(self, **kwargs: Unpack[K]) -> Generator[Tensor]:
+        predictions = self(**kwargs)[0]
+        labels = kwargs["labels"]
+
+        yield F.mse_loss(predictions, labels)
+
+    def metrics(self, **kwargs: Unpack[K]) -> dict[str, float]:
+        with torch.no_grad():
+            loss = next(self.loss(**kwargs)).item()
+
+            predictions = self(**kwargs)[0]
+            labels = kwargs["labels"]
+            mae = F.l1_loss(predictions, labels).item()
+
+        return {
+            "loss": loss,
+            "mse": loss,
+            "mae": mae,
+            "grad_norm": get_grad_norm(self)
+        }
+
+    def optimizers(
+        self,
+        **kwargs: Unpack[OptimizerKwargs],
+    ) -> tuple[Optimizer, ...]:
+        """
+        """
+
+        default_kwargs: Unpack[OptimizerKwargs] = {
+            "lr": 1e-3,
+            "eps": 1e-8,
+        }
+        opt_kwargs = {**default_kwargs, **kwargs}
+
+        optimizer = torch.optim.AdamW(
+            self.parameters(),
+            **opt_kwargs,
+        )
+
+        return (optimizer,)
+
+    def epoch_step(self) -> None:
+        return None
+
+    def epoch_write(
+        self,
+        writer: SummaryWriter,
+        step: int | None = None,
+        val: bool = False,
+        **kwargs: Unpack[K],
+    ) -> None:
+        return None
+
+    def log_arch(self) -> None:
+        super().log_arch()
+
+        logger.info(f"| > {self.input_dim=}")
+        logger.info(f"| > {self.hidden_dims=}")
+        logger.info(f"| > {self.output_dim=}")

trainlib/estimators/rnn.py

@@ -107,8 +107,14 @@ class LSTM[K: RNNKwargs](Estimator[K]):
         with torch.no_grad():
             loss = next(self.loss(**kwargs)).item()
 
+            predictions = self(**kwargs)[0]
+            labels = kwargs["labels"]
+            mae = F.l1_loss(predictions, labels).item()
+
         return {
             "loss": loss,
+            "mse": loss,
+            "mae": mae,
             "grad_norm": get_grad_norm(self)
         }
 
@@ -291,7 +297,7 @@ class MultiheadLSTM[K: MultiheadLSTMKwargs](Estimator[K]):
         logger.info(f"| > {self.output_dim=}")
 
 
-class ConvRNN[K: RNNKwargs](Estimator[K]):
+class ConvGRU[K: RNNKwargs](Estimator[K]):
     """
     Base recurrent convolutional architecture.
 
@@ -441,8 +447,14 @@ class ConvRNN[K: RNNKwargs](Estimator[K]):
         with torch.no_grad():
             loss = next(self.loss(**kwargs)).item()
 
+            predictions = self(**kwargs)[0].squeeze(-1)
+            labels = kwargs["labels"]
+            mae = F.l1_loss(predictions, labels).item()
+
         return {
             "loss": loss,
+            "mse": loss,
+            "mae": mae,
             "grad_norm": get_grad_norm(self)
         }
 

trainlib/trainer.py

@@ -1,3 +1,7 @@
+"""
+Core interface for training ``Estimators`` with ``Datasets``
+"""
+
 import os
 import time
 import logging
@@ -31,7 +35,15 @@ logger: logging.Logger = logging.getLogger(__name__)
 
 class Trainer[I, K: EstimatorKwargs]:
     """
-    Training interface for updating ``Estimators`` with ``Datasets``.
+    Training interface for optimizing parameters of ``Estimators`` with
+    ``Datasets``.
+
+    This class is generic to a dataset item type ``I`` and an estimator kwarg
+    type ``K``. These are the two primary components ``Trainer`` objects need
+    to coordinate: they ultimately rely on a provided map to ensure data items
+    (type ``I``) from a dataset are appropriately routed as inputs to key
+    estimator methods (like ``forward()`` and ``loss()``), which accept inputs
+    of type ``K``.
     """
 
     def __init__(
@@ -43,8 +55,10 @@ class Trainer[I, K: EstimatorKwargs]:
     ) -> None:
         """
         Parameters:
-            estimator: `Estimator` model object
+            estimator: ``Estimator`` model object
             device: device on which to carry out training
+            chkpt_dir: directory to write model checkpoints
+            tblog_dir: directory to write TensorBoard logs
         """
 
         self.device: str
@@ -87,7 +101,7 @@ class Trainer[I, K: EstimatorKwargs]:
 
     def reset(self) -> None:
         """
-        Set base tracking parameters.
+        Set initial tracking parameters for the primary training loop.
         """
 
         self._step: int = 0
@@ -244,10 +258,10 @@ class Trainer[I, K: EstimatorKwargs]:
 
         return val_loss_sums
 
-    def train(
+    def train[B](
         self,
         dataset: BatchedDataset[..., ..., I],
-        batch_estimator_map: Callable[[I, Self], K],
+        batch_estimator_map: Callable[[B, Self], K],
         lr: float = 1e-3,
         eps: float = 1e-8,
         max_grad_norm: float | None = None,
@@ -266,6 +280,10 @@ class Trainer[I, K: EstimatorKwargs]:
         summary_writer: SummaryWriter | None = None,
     ) -> Estimator:
         """
+        TODO: consider making the dataloader ``collate_fn`` an explicit
+        parameter with a type signature that reflects ``B``, connecting the
+        ``batch_estimator_map`` somewhere
+
         Note: this method attempts to implement a general scheme for passing
         needed items to the estimator's loss function from the dataloader. The
         abstract ``Estimator`` base only requires the model output be provided
@@ -276,13 +294,13 @@ class Trainer[I, K: EstimatorKwargs]:
         one should take care to synchronize the sample structure with `dataset`
         to match that expected by ``self.estimator.loss(...)``.
 
-        .. admonition:: On batch_estimator_map
+        .. admonition:: On ``batch_estimator_map``
 
             Dataloader collate functions are responsible for mapping a
             collection of items into an item of collections, roughly speaking.
             If items are tuples of tensors,
 
-            .. code-block::
+            .. code-block:: text
 
                 [
                     ( [1, 1], [1, 1] ),
@@ -293,7 +311,7 @@ class Trainer[I, K: EstimatorKwargs]:
             the collate function maps back into the item skeleton, producing a
             single tuple of (stacked) tensors
 
-            .. code-block::
+            .. code-block:: text
 
                 ( [[1, 1],
                    [2, 2],
@@ -303,14 +321,21 @@ class Trainer[I, K: EstimatorKwargs]:
                    [2, 2],
                    [3, 3]] )
 
-            This function should map from batches (which should be *item
+            This function should map from batches - which *may* be item
-            shaped*, i.e., have an ``I`` skeleton, even if stacked items may be
+            shaped, i.e., have an ``I`` skeleton, even if stacked items may be
-            different on the inside) into estimator keyword arguments (type
+            different on the inside - into estimator keyword arguments (type
-            ``K``).
+            ``K``). Collation behavior from a DataLoader (which can be
+            customized) doesn't consistently yield a known type shape, however,
+            so it's not appropriate to use ``I`` as the callable param type.
 
         Parameters:
+            dataset: dataset to train the estimator
+            batch_estimator_map: function mapping from batch data to expected
+                estimator kwargs
             lr: learning rate (default: 1e-3)
             eps: adam EPS (default: 1e-8)
+            max_grad_norm: upper bound to use when clipping gradients. If left
+                as ``None``, no gradient clipping is performed.
             max_epochs: maximum number of training epochs
             stop_after_epochs: number of epochs with stagnant validation losses
                 to allow before early stopping. If training stops earlier, the
@@ -395,11 +420,7 @@ class Trainer[I, K: EstimatorKwargs]:
 
             # save checkpoint
             if self._epoch % chkpt_every == 0:
-                self.save_model(
+                self.save_model(self._epoch, self.chkpt_dir, dir_prefix)
-                    self._epoch,
-                    self.chkpt_dir,
-                    dir_prefix
-                )
 
             self._epoch += 1
 
@@ -493,7 +514,7 @@ class Trainer[I, K: EstimatorKwargs]:
 
     def _summarize(self, writer: SummaryWriter, epoch: int) -> None:
         """
-        Flush the training summary to the TB summary writer.
+        Flush the training summary to the TensorBoard summary writer.
         """
 
         summary_values = defaultdict(list)
@@ -547,17 +568,18 @@ class Trainer[I, K: EstimatorKwargs]:
         chkpt_dir.mkdir(parents=True, exist_ok=True)
         chkpt_path.write_bytes(model_buff.getvalue())
 
-    def load_model(
+    def load_model(self, epoch: int, chkpt_dir: str) -> None:
-        self,
-        epoch: int,
-        chkpt_dir: str,
-    ) -> None:
         """
         Load a model checkpoint from a given epoch.
 
-        Note that this assumes the model was saved via `Trainer.save_model()`,
+        Note that this assumes the model was saved via
-        and the estimator provided to this `Trainer` instance matches the
+        ``Trainer.save_model()``, and the estimator provided to this
-        architecture of the checkpoint model being loaded.
+        ``Trainer`` instance matches the architecture of the checkpoint model
+        being loaded.
+
+        Parameters:
+            epoch: epoch of saved model
+            chkpt_dir: 
         """
 
         model_class = self.estimator.__class__.__name__

trainlib/transform.py

@@ -1,3 +1,7 @@
+"""
+Transform base for dataset records
+"""
+
 class Transform[I]:
     """
     Dataset transform base class.
@@ -8,4 +12,14 @@ class Transform[I]:
     """
 
     def __call__(self, item: I) -> I:
+        """
+        Apply transform to item.
+
+        Parameters:
+            item: item object to transform
+
+        Returns:
+            transformed item (same type ``I`` as input)
+        """
+
         raise NotImplementedError

trainlib/utils/custom.mplstyle

@@ -0,0 +1,46 @@
+text.usetex          : False
+mathtext.default     : regular
+
+font.family          : sans-serif
+font.sans-serif      : DejaVu Sans
+font.serif           : DejaVu Serif
+font.cursive         : DejaVu Sans
+mathtext.fontset     : dejavuserif
+font.size            : 9
+figure.titlesize     : 9
+legend.fontsize      : 9
+axes.titlesize       : 9
+axes.labelsize       : 9
+xtick.labelsize      : 9
+ytick.labelsize      : 9
+
+#axes.prop_cycle : cycler('color', ['4f7dd5', 'af7031', '55905e', 'd84739', '888348', 'b75e8b', '2f8f99', '9862cb'])
+axes.prop_cycle : cycler('color', ['5e8de4', 'c38141', '67a771', 'e15344', '9e9858', '41a6b0', 'a46fd7', 'c86d9a'])
+
+image.interpolation   : nearest
+image.resample        : False
+image.composite_image : True
+
+axes.spines.left     : True
+axes.spines.bottom   : True
+axes.spines.top      : False
+axes.spines.right    : False
+
+axes.linewidth       : 1
+xtick.major.width    : 1
+xtick.minor.width    : 1
+ytick.major.width    : 1
+ytick.minor.width    : 1
+
+lines.linewidth      : 1
+lines.markersize     : 1
+
+savefig.dpi          : 300
+savefig.format       : svg
+savefig.bbox         : tight
+savefig.pad_inches   : 0.1
+
+svg.image_inline     : True
+svg.fonttype         : none
+
+legend.frameon       : False

trainlib/utils/plot.py

@@ -0,0 +1,38 @@
+from pathlib import Path
+
+import matplotlib as mpl
+import matplotlib.pyplot as plt
+
+FILE = Path(__file__).parent.absolute()
+
+
+class use_style:
+    def __init__(
+        self,
+        style: list[str] | None = None,
+        **kwargs: str,
+    ) -> None:
+        super().__init__()
+
+        if style is None:
+            style = [str(Path(FILE, "custom.mplstyle"))]
+        self.style = style + [kwargs]
+        self.previous_style = {}
+
+    def __enter__(self) -> None:
+        self.previous_style = mpl.rcParams.copy()
+        if self.style is not None:
+            plt.style.use(self.style)
+
+    def __exit__(self, *args: str, **kwargs: str) -> None:
+        mpl.rcParams.update(self.previous_style)
+
+
+def set_style(
+    style: list[str] | None = None,
+    **kwargs: str,
+) -> None:
+    if style is None:
+        style = [str(Path(FILE, "custom.mplstyle"))]
+
+    plt.style.use(style + [kwargs])

trainlib/utils/session.py

@@ -0,0 +1,21 @@
+import random
+
+import numpy as np
+import torch
+from torch import Tensor
+
+
+def seed_all_backends(seed: int | Tensor | None = None) -> None:
+    """Sets all python, numpy and pytorch seeds."""
+
+    if seed is None:
+        seed = int(torch.randint(1000000, size=(1,)))
+    else:
+        seed = int(seed)
+
+    random.seed(seed)
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False

uv.lock

@@ -248,9 +248,13 @@ dependencies = [
     { name = "cuda-pathfinder" },
 ]
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/05/8b/b4b2d1c7775fa403b64333e720cfcfccef8dcb9cdeb99947061ca5a77628/cuda_bindings-12.9.4-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:cf8bfaedc238f3b115d957d1fd6562b7e8435ba57f6d0e2f87d0e7149ccb2da5", size = 11570071, upload-time = "2025-10-21T14:51:47.472Z" },
     { url = "https://files.pythonhosted.org/packages/63/56/e465c31dc9111be3441a9ba7df1941fe98f4aa6e71e8788a3fb4534ce24d/cuda_bindings-12.9.4-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:32bdc5a76906be4c61eb98f546a6786c5773a881f3b166486449b5d141e4a39f", size = 11906628, upload-time = "2025-10-21T14:51:49.905Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/07/6aff13bc1e977e35aaa6b22f52b172e2890c608c6db22438cf7ed2bf43a6/cuda_bindings-12.9.4-cp313-cp313t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3adf4958dcf68ae7801a59b73fb00a8b37f8d0595060d66ceae111b1002de38d", size = 11566797, upload-time = "2025-10-21T14:51:54.581Z" },
     { url = "https://files.pythonhosted.org/packages/a3/84/1e6be415e37478070aeeee5884c2022713c1ecc735e6d82d744de0252eee/cuda_bindings-12.9.4-cp313-cp313t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:56e0043c457a99ac473ddc926fe0dc4046694d99caef633e92601ab52cbe17eb", size = 11925991, upload-time = "2025-10-21T14:51:56.535Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/b5/96a6696e20c4ffd2b327f54c7d0fde2259bdb998d045c25d5dedbbe30290/cuda_bindings-12.9.4-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1f53a7f453d4b2643d8663d036bafe29b5ba89eb904c133180f295df6dc151e5", size = 11624530, upload-time = "2025-10-21T14:52:01.539Z" },
     { url = "https://files.pythonhosted.org/packages/d1/af/6dfd8f2ed90b1d4719bc053ff8940e494640fe4212dc3dd72f383e4992da/cuda_bindings-12.9.4-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8b72ee72a9cc1b531db31eebaaee5c69a8ec3500e32c6933f2d3b15297b53686", size = 11922703, upload-time = "2025-10-21T14:52:03.585Z" },
+    { url = "https://files.pythonhosted.org/packages/39/73/d2fc40c043bac699c3880bf88d3cebe9d88410cd043795382826c93a89f0/cuda_bindings-12.9.4-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:20f2699d61d724de3eb3f3369d57e2b245f93085cab44fd37c3bea036cea1a6f", size = 11565056, upload-time = "2025-10-21T14:52:08.338Z" },
     { url = "https://files.pythonhosted.org/packages/6c/19/90ac264acc00f6df8a49378eedec9fd2db3061bf9263bf9f39fd3d8377c3/cuda_bindings-12.9.4-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d80bffc357df9988dca279734bc9674c3934a654cab10cadeed27ce17d8635ee", size = 11924658, upload-time = "2025-10-21T14:52:10.411Z" },
 ]
 
@@ -861,6 +865,7 @@ name = "nvidia-cublas-cu12"
 version = "12.8.4.1"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/29/99/db44d685f0e257ff0e213ade1964fc459b4a690a73293220e98feb3307cf/nvidia_cublas_cu12-12.8.4.1-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:b86f6dd8935884615a0683b663891d43781b819ac4f2ba2b0c9604676af346d0", size = 590537124, upload-time = "2025-03-07T01:43:53.556Z" },
     { url = "https://files.pythonhosted.org/packages/dc/61/e24b560ab2e2eaeb3c839129175fb330dfcfc29e5203196e5541a4c44682/nvidia_cublas_cu12-12.8.4.1-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:8ac4e771d5a348c551b2a426eda6193c19aa630236b418086020df5ba9667142", size = 594346921, upload-time = "2025-03-07T01:44:31.254Z" },
 ]
 
@@ -869,6 +874,7 @@ name = "nvidia-cuda-cupti-cu12"
 version = "12.8.90"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/d5/1f/b3bd73445e5cb342727fd24fe1f7b748f690b460acadc27ea22f904502c8/nvidia_cuda_cupti_cu12-12.8.90-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:4412396548808ddfed3f17a467b104ba7751e6b58678a4b840675c56d21cf7ed", size = 9533318, upload-time = "2025-03-07T01:40:10.421Z" },
     { url = "https://files.pythonhosted.org/packages/f8/02/2adcaa145158bf1a8295d83591d22e4103dbfd821bcaf6f3f53151ca4ffa/nvidia_cuda_cupti_cu12-12.8.90-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ea0cb07ebda26bb9b29ba82cda34849e73c166c18162d3913575b0c9db9a6182", size = 10248621, upload-time = "2025-03-07T01:40:21.213Z" },
 ]
 
@@ -878,6 +884,7 @@ version = "12.8.93"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/05/6b/32f747947df2da6994e999492ab306a903659555dddc0fbdeb9d71f75e52/nvidia_cuda_nvrtc_cu12-12.8.93-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:a7756528852ef889772a84c6cd89d41dfa74667e24cca16bb31f8f061e3e9994", size = 88040029, upload-time = "2025-03-07T01:42:13.562Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/d1/e50d0acaab360482034b84b6e27ee83c6738f7d32182b987f9c7a4e32962/nvidia_cuda_nvrtc_cu12-12.8.93-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:fc1fec1e1637854b4c0a65fb9a8346b51dd9ee69e61ebaccc82058441f15bce8", size = 43106076, upload-time = "2025-03-07T01:41:59.817Z" },
 ]
 
 [[package]]
@@ -885,6 +892,7 @@ name = "nvidia-cuda-runtime-cu12"
 version = "12.8.90"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/7c/75/f865a3b236e4647605ea34cc450900854ba123834a5f1598e160b9530c3a/nvidia_cuda_runtime_cu12-12.8.90-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:52bf7bbee900262ffefe5e9d5a2a69a30d97e2bc5bb6cc866688caa976966e3d", size = 965265, upload-time = "2025-03-07T01:39:43.533Z" },
     { url = "https://files.pythonhosted.org/packages/0d/9b/a997b638fcd068ad6e4d53b8551a7d30fe8b404d6f1804abf1df69838932/nvidia_cuda_runtime_cu12-12.8.90-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:adade8dcbd0edf427b7204d480d6066d33902cab2a4707dcfc48a2d0fd44ab90", size = 954765, upload-time = "2025-03-07T01:40:01.615Z" },
 ]
 
@@ -896,6 +904,7 @@ dependencies = [
     { name = "nvidia-cublas-cu12" },
 ]
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/fa/41/e79269ce215c857c935fd86bcfe91a451a584dfc27f1e068f568b9ad1ab7/nvidia_cudnn_cu12-9.10.2.21-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:c9132cc3f8958447b4910a1720036d9eff5928cc3179b0a51fb6d167c6cc87d8", size = 705026878, upload-time = "2025-06-06T21:52:51.348Z" },
     { url = "https://files.pythonhosted.org/packages/ba/51/e123d997aa098c61d029f76663dedbfb9bc8dcf8c60cbd6adbe42f76d049/nvidia_cudnn_cu12-9.10.2.21-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:949452be657fa16687d0930933f032835951ef0892b37d2d53824d1a84dc97a8", size = 706758467, upload-time = "2025-06-06T21:54:08.597Z" },
 ]
 
@@ -907,6 +916,7 @@ dependencies = [
     { name = "nvidia-nvjitlink-cu12" },
 ]
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/60/bc/7771846d3a0272026c416fbb7e5f4c1f146d6d80704534d0b187dd6f4800/nvidia_cufft_cu12-11.3.3.83-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:848ef7224d6305cdb2a4df928759dca7b1201874787083b6e7550dd6765ce69a", size = 193109211, upload-time = "2025-03-07T01:44:56.873Z" },
     { url = "https://files.pythonhosted.org/packages/1f/13/ee4e00f30e676b66ae65b4f08cb5bcbb8392c03f54f2d5413ea99a5d1c80/nvidia_cufft_cu12-11.3.3.83-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:4d2dd21ec0b88cf61b62e6b43564355e5222e4a3fb394cac0db101f2dd0d4f74", size = 193118695, upload-time = "2025-03-07T01:45:27.821Z" },
 ]
 
@@ -916,6 +926,7 @@ version = "1.13.1.3"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/bb/fe/1bcba1dfbfb8d01be8d93f07bfc502c93fa23afa6fd5ab3fc7c1df71038a/nvidia_cufile_cu12-1.13.1.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:1d069003be650e131b21c932ec3d8969c1715379251f8d23a1860554b1cb24fc", size = 1197834, upload-time = "2025-03-07T01:45:50.723Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/f5/5607710447a6fe9fd9b3283956fceeee8a06cda1d2f56ce31371f595db2a/nvidia_cufile_cu12-1.13.1.3-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:4beb6d4cce47c1a0f1013d72e02b0994730359e17801d395bdcbf20cfb3bb00a", size = 1120705, upload-time = "2025-03-07T01:45:41.434Z" },
 ]
 
 [[package]]
@@ -923,6 +934,7 @@ name = "nvidia-curand-cu12"
 version = "10.3.9.90"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/45/5e/92aa15eca622a388b80fbf8375d4760738df6285b1e92c43d37390a33a9a/nvidia_curand_cu12-10.3.9.90-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:dfab99248034673b779bc6decafdc3404a8a6f502462201f2f31f11354204acd", size = 63625754, upload-time = "2025-03-07T01:46:10.735Z" },
     { url = "https://files.pythonhosted.org/packages/fb/aa/6584b56dc84ebe9cf93226a5cde4d99080c8e90ab40f0c27bda7a0f29aa1/nvidia_curand_cu12-10.3.9.90-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:b32331d4f4df5d6eefa0554c565b626c7216f87a06a4f56fab27c3b68a830ec9", size = 63619976, upload-time = "2025-03-07T01:46:23.323Z" },
 ]
 
@@ -936,6 +948,7 @@ dependencies = [
     { name = "nvidia-nvjitlink-cu12" },
 ]
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/c8/32/f7cd6ce8a7690544d084ea21c26e910a97e077c9b7f07bf5de623ee19981/nvidia_cusolver_cu12-11.7.3.90-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:db9ed69dbef9715071232caa9b69c52ac7de3a95773c2db65bdba85916e4e5c0", size = 267229841, upload-time = "2025-03-07T01:46:54.356Z" },
     { url = "https://files.pythonhosted.org/packages/85/48/9a13d2975803e8cf2777d5ed57b87a0b6ca2cc795f9a4f59796a910bfb80/nvidia_cusolver_cu12-11.7.3.90-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:4376c11ad263152bd50ea295c05370360776f8c3427b30991df774f9fb26c450", size = 267506905, upload-time = "2025-03-07T01:47:16.273Z" },
 ]
 
@@ -947,6 +960,7 @@ dependencies = [
     { name = "nvidia-nvjitlink-cu12" },
 ]
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/bc/f7/cd777c4109681367721b00a106f491e0d0d15cfa1fd59672ce580ce42a97/nvidia_cusparse_cu12-12.5.8.93-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:9b6c161cb130be1a07a27ea6923df8141f3c295852f4b260c65f18f3e0a091dc", size = 288117129, upload-time = "2025-03-07T01:47:40.407Z" },
     { url = "https://files.pythonhosted.org/packages/c2/f5/e1854cb2f2bcd4280c44736c93550cc300ff4b8c95ebe370d0aa7d2b473d/nvidia_cusparse_cu12-12.5.8.93-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:1ec05d76bbbd8b61b06a80e1eaf8cf4959c3d4ce8e711b65ebd0443bb0ebb13b", size = 288216466, upload-time = "2025-03-07T01:48:13.779Z" },
 ]
 
@@ -955,6 +969,7 @@ name = "nvidia-cusparselt-cu12"
 version = "0.7.1"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/73/b9/598f6ff36faaece4b3c50d26f50e38661499ff34346f00e057760b35cc9d/nvidia_cusparselt_cu12-0.7.1-py3-none-manylinux2014_aarch64.whl", hash = "sha256:8878dce784d0fac90131b6817b607e803c36e629ba34dc5b433471382196b6a5", size = 283835557, upload-time = "2025-02-26T00:16:54.265Z" },
     { url = "https://files.pythonhosted.org/packages/56/79/12978b96bd44274fe38b5dde5cfb660b1d114f70a65ef962bcbbed99b549/nvidia_cusparselt_cu12-0.7.1-py3-none-manylinux2014_x86_64.whl", hash = "sha256:f1bb701d6b930d5a7cea44c19ceb973311500847f81b634d802b7b539dc55623", size = 287193691, upload-time = "2025-02-26T00:15:44.104Z" },
 ]
 
@@ -963,6 +978,7 @@ name = "nvidia-nccl-cu12"
 version = "2.27.5"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/bb/1c/857979db0ef194ca5e21478a0612bcdbbe59458d7694361882279947b349/nvidia_nccl_cu12-2.27.5-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:31432ad4d1fb1004eb0c56203dc9bc2178a1ba69d1d9e02d64a6938ab5e40e7a", size = 322400625, upload-time = "2025-06-26T04:11:04.496Z" },
     { url = "https://files.pythonhosted.org/packages/6e/89/f7a07dc961b60645dbbf42e80f2bc85ade7feb9a491b11a1e973aa00071f/nvidia_nccl_cu12-2.27.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ad730cf15cb5d25fe849c6e6ca9eb5b76db16a80f13f425ac68d8e2e55624457", size = 322348229, upload-time = "2025-06-26T04:11:28.385Z" },
 ]
 
@@ -972,6 +988,7 @@ version = "12.8.93"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/f6/74/86a07f1d0f42998ca31312f998bd3b9a7eff7f52378f4f270c8679c77fb9/nvidia_nvjitlink_cu12-12.8.93-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:81ff63371a7ebd6e6451970684f916be2eab07321b73c9d244dc2b4da7f73b88", size = 39254836, upload-time = "2025-03-07T01:49:55.661Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/a2/8cee5da30d13430e87bf99bb33455d2724d0a4a9cb5d7926d80ccb96d008/nvidia_nvjitlink_cu12-12.8.93-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:adccd7161ace7261e01bb91e44e88da350895c270d23f744f0820c818b7229e7", size = 38386204, upload-time = "2025-03-07T01:49:43.612Z" },
 ]
 
 [[package]]
@@ -979,6 +996,7 @@ name = "nvidia-nvshmem-cu12"
 version = "3.4.5"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/1d/6a/03aa43cc9bd3ad91553a88b5f6fb25ed6a3752ae86ce2180221962bc2aa5/nvidia_nvshmem_cu12-3.4.5-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:0b48363fc6964dede448029434c6abed6c5e37f823cb43c3bcde7ecfc0457e15", size = 138936938, upload-time = "2025-09-06T00:32:05.589Z" },
     { url = "https://files.pythonhosted.org/packages/b5/09/6ea3ea725f82e1e76684f0708bbedd871fc96da89945adeba65c3835a64c/nvidia_nvshmem_cu12-3.4.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:042f2500f24c021db8a06c5eec2539027d57460e1c1a762055a6554f72c369bd", size = 139103095, upload-time = "2025-09-06T00:32:31.266Z" },
 ]
 
@@ -987,6 +1005,7 @@ name = "nvidia-nvtx-cu12"
 version = "12.8.90"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/10/c0/1b303feea90d296f6176f32a2a70b5ef230f9bdeb3a72bddb0dc922dc137/nvidia_nvtx_cu12-12.8.90-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d7ad891da111ebafbf7e015d34879f7112832fc239ff0d7d776b6cb685274615", size = 91161, upload-time = "2025-03-07T01:42:23.922Z" },
     { url = "https://files.pythonhosted.org/packages/a2/eb/86626c1bbc2edb86323022371c39aa48df6fd8b0a1647bc274577f72e90b/nvidia_nvtx_cu12-12.8.90-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:5b17e2001cc0d751a5bc2c6ec6d26ad95913324a4adb86788c944f8ce9ba441f", size = 89954, upload-time = "2025-03-07T01:42:44.131Z" },
 ]
 
@@ -1393,14 +1412,14 @@ wheels = [
 
 [[package]]
 name = "sphinx-autodoc-typehints"
-version = "3.9.5"
+version = "3.9.7"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "sphinx" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/58/ec/21bd9babcfeb9930a73011257002d5cfa5fd30667b8de6d76dbaf8275dfb/sphinx_autodoc_typehints-3.9.5.tar.gz", hash = "sha256:60e646efb7c352a0e98f34dd7fdcde4527fbbdbdf30371ff8321b6b3eb1fd37d", size = 63249, upload-time = "2026-03-02T19:58:07.974Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/f4/06/da2d9e98b3f7f0df144496e62f453e0025f129bccc7a6076b8ceae6047b1/sphinx_autodoc_typehints-3.9.7.tar.gz", hash = "sha256:70f3dd4e4dd815ae30e5d3848a26dca71fb5e7fcf8f37cf8b840dc8afdf07e82", size = 68689, upload-time = "2026-03-05T18:33:40.829Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/7f/cb/80c250f47a0ca5ac67d82f14811b4068a551a12b4790b085ffdb900de427/sphinx_autodoc_typehints-3.9.5-py3-none-any.whl", hash = "sha256:c94f88a90b6c61a7a6686cb77b410e46e077712838387e6cf22d69e85cfd06a5", size = 34763, upload-time = "2026-03-02T19:58:06.028Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/a0/e7d3365dabfa79a1b2ac7d3122b5b22b401a9c4d5e4eadc5e13b88c63a2c/sphinx_autodoc_typehints-3.9.7-py3-none-any.whl", hash = "sha256:dd73f6a32adef0d8208f6f7d99254e1880259c77db7b4a91648345d45202d48e", size = 36691, upload-time = "2026-03-05T18:33:38.983Z" },
 ]
 
 [[package]]
@@ -1542,52 +1561,47 @@ wheels = [
 
 [[package]]
 name = "torch"
-version = "2.10.0"
+version = "2.10.0+cu128"
-source = { registry = "https://pypi.org/simple" }
+source = { registry = "https://download.pytorch.org/whl/cu128" }
 dependencies = [
-    { name = "cuda-bindings", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "cuda-bindings", marker = "sys_platform == 'linux'" },
     { name = "filelock" },
     { name = "fsspec" },
     { name = "jinja2" },
     { name = "networkx" },
-    { name = "nvidia-cublas-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cublas-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-cuda-cupti-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-cupti-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-cuda-nvrtc-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-nvrtc-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-cuda-runtime-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-runtime-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-cudnn-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cudnn-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-cufft-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufft-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-cufile-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufile-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-curand-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-curand-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-cusolver-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusolver-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-cusparse-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparse-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-cusparselt-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparselt-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-nccl-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nccl-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-nvjitlink-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvjitlink-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-nvshmem-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvshmem-cu12", marker = "sys_platform == 'linux'" },
-    { name = "nvidia-nvtx-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvtx-cu12", marker = "sys_platform == 'linux'" },
     { name = "setuptools" },
     { name = "sympy" },
-    { name = "triton", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "triton", marker = "sys_platform == 'linux'" },
     { name = "typing-extensions" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ec/23/2c9fe0c9c27f7f6cb865abcea8a4568f29f00acaeadfc6a37f6801f84cb4/torch-2.10.0-2-cp313-none-macosx_11_0_arm64.whl", hash = "sha256:e521c9f030a3774ed770a9c011751fb47c4d12029a3d6522116e48431f2ff89e", size = 79498254, upload-time = "2026-02-10T21:44:44.095Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:bdbcc703382f948e951c063448c9406bf38ce66c41dd698d9e2733fcf96c037a" },
-    { url = "https://files.pythonhosted.org/packages/c9/6f/f2e91e34e3fcba2e3fc8d8f74e7d6c22e74e480bbd1db7bc8900fdf3e95c/torch-2.10.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:5c4d217b14741e40776dd7074d9006fd28b8a97ef5654db959d8635b2fe5f29b", size = 146004247, upload-time = "2026-01-21T16:24:29.335Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:7b4bd23ed63de97456fcc81c26fea9f02ee02ce1112111c4dac0d8cfe574b23e" },
-    { url = "https://files.pythonhosted.org/packages/98/fb/5160261aeb5e1ee12ee95fe599d0541f7c976c3701d607d8fc29e623229f/torch-2.10.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:6b71486353fce0f9714ca0c9ef1c850a2ae766b409808acd58e9678a3edb7738", size = 915716445, upload-time = "2026-01-21T16:22:45.353Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp313-cp313-win_amd64.whl", hash = "sha256:4d1b0b49c54223c7c04050b49eac141d77b6edbc34aea1dfc74a6fdb661baa8c" },
-    { url = "https://files.pythonhosted.org/packages/6a/16/502fb1b41e6d868e8deb5b0e3ae926bbb36dab8ceb0d1b769b266ad7b0c3/torch-2.10.0-cp313-cp313-win_amd64.whl", hash = "sha256:c2ee399c644dc92ef7bc0d4f7e74b5360c37cdbe7c5ba11318dda49ffac2bc57", size = 113757050, upload-time = "2026-01-21T16:24:19.204Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:f1f8b840c64b645a4bc61a393db48effb9c92b2dc26c8373873911f0750d1ea7" },
-    { url = "https://files.pythonhosted.org/packages/1a/0b/39929b148f4824bc3ad6f9f72a29d4ad865bcf7ebfc2fa67584773e083d2/torch-2.10.0-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:3202429f58309b9fa96a614885eace4b7995729f44beb54d3e4a47773649d382", size = 79851305, upload-time = "2026-01-21T16:24:09.209Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:23f58258012bcf1c349cb22af387e33aadca7f83ea617b080e774eb41e4fe8ff" },
-    { url = "https://files.pythonhosted.org/packages/d8/14/21fbce63bc452381ba5f74a2c0a959fdf5ad5803ccc0c654e752e0dbe91a/torch-2.10.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:aae1b29cd68e50a9397f5ee897b9c24742e9e306f88a807a27d617f07adb3bd8", size = 146005472, upload-time = "2026-01-21T16:22:29.022Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp313-cp313t-win_amd64.whl", hash = "sha256:01b216e097b17a5277cfb47c383cdcacf06abeadcb0daca0c76b59e72854c3b6" },
-    { url = "https://files.pythonhosted.org/packages/54/fd/b207d1c525cb570ef47f3e9f836b154685011fce11a2f444ba8a4084d042/torch-2.10.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:6021db85958db2f07ec94e1bc77212721ba4920c12a18dc552d2ae36a3eb163f", size = 915612644, upload-time = "2026-01-21T16:21:47.019Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp314-cp314-manylinux_2_28_aarch64.whl", hash = "sha256:c42377bc2607e3e1c60da71b792fb507c3938c87fd6edab8b21c59c91473c36d" },
-    { url = "https://files.pythonhosted.org/packages/36/53/0197f868c75f1050b199fe58f9bf3bf3aecac9b4e85cc9c964383d745403/torch-2.10.0-cp313-cp313t-win_amd64.whl", hash = "sha256:ff43db38af76fda183156153983c9a096fc4c78d0cd1e07b14a2314c7f01c2c8", size = 113997015, upload-time = "2026-01-21T16:23:00.767Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp314-cp314-manylinux_2_28_x86_64.whl", hash = "sha256:37d71feea068776855686a1512058df3f19f6f040a151f055aa746601678744f" },
-    { url = "https://files.pythonhosted.org/packages/0e/13/e76b4d9c160e89fff48bf16b449ea324bda84745d2ab30294c37c2434c0d/torch-2.10.0-cp313-none-macosx_11_0_arm64.whl", hash = "sha256:cdf2a523d699b70d613243211ecaac14fe9c5df8a0b0a9c02add60fb2a413e0f", size = 79498248, upload-time = "2026-01-21T16:23:09.315Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp314-cp314-win_amd64.whl", hash = "sha256:c57017ca29e62271e362fdeee7d20070e254755a5148b30b553d8a10fc83c7ef" },
-    { url = "https://files.pythonhosted.org/packages/4f/93/716b5ac0155f1be70ed81bacc21269c3ece8dba0c249b9994094110bfc51/torch-2.10.0-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:bf0d9ff448b0218e0433aeb198805192346c4fd659c852370d5cc245f602a06a", size = 79464992, upload-time = "2026-01-21T16:23:05.162Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:777461f50b2daf77e4bdd8e2ad34bdfc5a993bf1bdf2ab9ef39f5edfe4e9c12b" },
-    { url = "https://files.pythonhosted.org/packages/69/2b/51e663ff190c9d16d4a8271203b71bc73a16aa7619b9f271a69b9d4a936b/torch-2.10.0-cp314-cp314-manylinux_2_28_aarch64.whl", hash = "sha256:233aed0659a2503b831d8a67e9da66a62c996204c0bba4f4c442ccc0c68a3f60", size = 146018567, upload-time = "2026-01-21T16:22:23.393Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:7bcba6a7c5f0987a13298b1ca843155dcceceac758fa3c7ccd5c7af4059a1080" },
-    { url = "https://files.pythonhosted.org/packages/5e/cd/4b95ef7f293b927c283db0b136c42be91c8ec6845c44de0238c8c23bdc80/torch-2.10.0-cp314-cp314-manylinux_2_28_x86_64.whl", hash = "sha256:682497e16bdfa6efeec8cde66531bc8d1fbbbb4d8788ec6173c089ed3cc2bfe5", size = 915721646, upload-time = "2026-01-21T16:21:16.983Z" },
+    { url = "https://download.pytorch.org/whl/cu128/torch-2.10.0%2Bcu128-cp314-cp314t-win_amd64.whl", hash = "sha256:70d89143c956389d4806cb4e5fe0b1129fe0db280e1073288d17fa76c101cba4" },
-    { url = "https://files.pythonhosted.org/packages/56/97/078a007208f8056d88ae43198833469e61a0a355abc0b070edd2c085eb9a/torch-2.10.0-cp314-cp314-win_amd64.whl", hash = "sha256:6528f13d2a8593a1a412ea07a99812495bec07e9224c28b2a25c0a30c7da025c", size = 113752373, upload-time = "2026-01-21T16:22:13.471Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/94/71994e7d0d5238393df9732fdab607e37e2b56d26a746cb59fdb415f8966/torch-2.10.0-cp314-cp314t-macosx_14_0_arm64.whl", hash = "sha256:f5ab4ba32383061be0fb74bda772d470140a12c1c3b58a0cfbf3dae94d164c28", size = 79850324, upload-time = "2026-01-21T16:22:09.494Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/65/1a05346b418ea8ccd10360eef4b3e0ce688fba544e76edec26913a8d0ee0/torch-2.10.0-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:716b01a176c2a5659c98f6b01bf868244abdd896526f1c692712ab36dbaf9b63", size = 146006482, upload-time = "2026-01-21T16:22:18.42Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/b9/5f6f9d9e859fc3235f60578fa64f52c9c6e9b4327f0fe0defb6de5c0de31/torch-2.10.0-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:d8f5912ba938233f86361e891789595ff35ca4b4e2ac8fe3670895e5976731d6", size = 915613050, upload-time = "2026-01-21T16:20:49.035Z" },
-    { url = "https://files.pythonhosted.org/packages/66/4d/35352043ee0eaffdeff154fad67cd4a31dbed7ff8e3be1cc4549717d6d51/torch-2.10.0-cp314-cp314t-win_amd64.whl", hash = "sha256:71283a373f0ee2c89e0f0d5f446039bdabe8dbc3c9ccf35f0f784908b0acd185", size = 113995816, upload-time = "2026-01-21T16:22:05.312Z" },
 ]
 
 [[package]]
@@ -1623,7 +1637,7 @@ wheels = [
 
 [[package]]
 name = "trainlib"
-version = "0.1.0"
+version = "0.1.2"
 source = { editable = "." }
 dependencies = [
     { name = "colorama" },
@@ -1662,7 +1676,7 @@ requires-dist = [
     { name = "sphinx-autodoc-typehints", marker = "extra == 'doc'" },
     { name = "sphinx-togglebutton", marker = "extra == 'doc'" },
     { name = "tensorboard", specifier = ">=2.20.0" },
-    { name = "torch", specifier = ">=2.5.1" },
+    { name = "torch", index = "https://download.pytorch.org/whl/cu128" },
     { name = "tqdm", specifier = ">=4.67.1" },
 ]
 provides-extras = ["dev", "doc", "test"]
@@ -1681,9 +1695,13 @@ name = "triton"
 version = "3.6.0"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
+    { url = "https://files.pythonhosted.org/packages/3c/12/34d71b350e89a204c2c7777a9bba0dcf2f19a5bfdd70b57c4dbc5ffd7154/triton-3.6.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:448e02fe6dc898e9e5aa89cf0ee5c371e99df5aa5e8ad976a80b93334f3494fd", size = 176133521, upload-time = "2026-01-20T16:16:13.321Z" },
     { url = "https://files.pythonhosted.org/packages/f9/0b/37d991d8c130ce81a8728ae3c25b6e60935838e9be1b58791f5997b24a54/triton-3.6.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:10c7f76c6e72d2ef08df639e3d0d30729112f47a56b0c81672edc05ee5116ac9", size = 188289450, upload-time = "2026-01-20T16:00:49.136Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/4e/41b0c8033b503fd3cfcd12392cdd256945026a91ff02452bef40ec34bee7/triton-3.6.0-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1722e172d34e32abc3eb7711d0025bb69d7959ebea84e3b7f7a341cd7ed694d6", size = 176276087, upload-time = "2026-01-20T16:16:18.989Z" },
     { url = "https://files.pythonhosted.org/packages/35/f8/9c66bfc55361ec6d0e4040a0337fb5924ceb23de4648b8a81ae9d33b2b38/triton-3.6.0-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d002e07d7180fd65e622134fbd980c9a3d4211fb85224b56a0a0efbd422ab72f", size = 188400296, upload-time = "2026-01-20T16:00:56.042Z" },
+    { url = "https://files.pythonhosted.org/packages/49/55/5ecf0dcaa0f2fbbd4420f7ef227ee3cb172e91e5fede9d0ecaddc43363b4/triton-3.6.0-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ef5523241e7d1abca00f1d240949eebdd7c673b005edbbce0aca95b8191f1d43", size = 176138577, upload-time = "2026-01-20T16:16:25.426Z" },
     { url = "https://files.pythonhosted.org/packages/df/3d/9e7eee57b37c80cec63322c0231bb6da3cfe535a91d7a4d64896fcb89357/triton-3.6.0-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a17a5d5985f0ac494ed8a8e54568f092f7057ef60e1b0fa09d3fd1512064e803", size = 188273063, upload-time = "2026-01-20T16:01:07.278Z" },
+    { url = "https://files.pythonhosted.org/packages/48/db/56ee649cab5eaff4757541325aca81f52d02d4a7cd3506776cad2451e060/triton-3.6.0-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0b3a97e8ed304dfa9bd23bb41ca04cdf6b2e617d5e782a8653d616037a5d537d", size = 176274804, upload-time = "2026-01-20T16:16:31.528Z" },
     { url = "https://files.pythonhosted.org/packages/f6/56/6113c23ff46c00aae423333eb58b3e60bdfe9179d542781955a5e1514cb3/triton-3.6.0-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:46bd1c1af4b6704e554cad2eeb3b0a6513a980d470ccfa63189737340c7746a7", size = 188397994, upload-time = "2026-01-20T16:01:14.236Z" },
 ]