update doc stable references for 0.1.4 release

CHANGELOG.md

@@ -8,18 +8,18 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ### Added
 
-- Added LR scheduler reinitialization functionality ([#2](https://github.com/speediedan/finetuning-scheduler/pull/2))
-- Added advanced usage documentation
-- Added advanced scheduling examples
-- added notebook-based tutorial link
+- LR scheduler reinitialization functionality ([#2](https://github.com/speediedan/finetuning-scheduler/pull/2))
+- advanced usage documentation
+- advanced scheduling examples
+- notebook-based tutorial link
 - enhanced cli-based example hparam logging among other code clarifications
 
-### Changed
-
 ### Fixed
 
 - addressed URI length limit for custom badge
 - allow new deberta fast tokenizer conversion warning for transformers >= 4.19
+### Changed
+
 ### Deprecated
 
 ## [0.1.3] - 2022-05-04

CITATION.cff

@@ -6,7 +6,7 @@ date-released: 2022-02-04
 authors:
   - family-names: "Dale"
     given-names: "Dan"
-version: 0.1.3
+version: 0.1.4
 identifiers:
   - description: "Finetuning Scheduler (all versions)"
     type: doi

README.md

@@ -7,7 +7,7 @@
 ______________________________________________________________________
 
 <p align="center">
-  <a href="https://finetuning-scheduler.readthedocs.io/en/latest/">Docs</a> •
+  <a href="https://finetuning-scheduler.readthedocs.io/en/stable/">Docs</a> •
   <a href="#Setup">Setup</a> •
   <a href="#examples">Examples</a> •
   <a href="#community">Community</a>
@@ -17,7 +17,7 @@ ______________________________________________________________________
 [![PyPI Status](https://badge.fury.io/py/finetuning-scheduler.svg)](https://badge.fury.io/py/finetuning-scheduler)
 ![Conda (channel only)](https://img.shields.io/conda/vn/conda-forge/finetuning-scheduler?color=%23000080)\
 [![codecov](https://codecov.io/gh/speediedan/finetuning-scheduler/branch/main/graph/badge.svg)](https://codecov.io/gh/speediedan/finetuning-scheduler)
-[![ReadTheDocs](https://readthedocs.org/projects/finetuning-scheduler/badge/?version=latest)](https://finetuning-scheduler.readthedocs.io/en/latest/)
+[![ReadTheDocs](https://readthedocs.org/projects/finetuning-scheduler/badge/?version=latest)](https://finetuning-scheduler.readthedocs.io/en/stable/)
 [![DOI](https://zenodo.org/badge/455666112.svg)](https://zenodo.org/badge/latestdoi/455666112)
 [![license](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://github.com/speediedan/finetuning-scheduler/blob/master/LICENSE)
 
@@ -27,7 +27,7 @@ ______________________________________________________________________
 
 <img width="300px" src="docs/source/_static/images/fts/fts_explicit_loss_anim.gif" alt="FinetuningScheduler explicit loss animation" align="right"/>
 
-[FinetuningScheduler](https://finetuning-scheduler.readthedocs.io/en/latest/api/finetuning_scheduler.fts.html#finetuning_scheduler.fts.FinetuningScheduler) is simple to use yet powerful, offering a number of features that facilitate model research and exploration:
+[FinetuningScheduler](https://finetuning-scheduler.readthedocs.io/en/stable/api/finetuning_scheduler.fts.html#finetuning_scheduler.fts.FinetuningScheduler) is simple to use yet powerful, offering a number of features that facilitate model research and exploration:
 
 - easy specification of flexible finetuning schedules with explicit or regex-based parameter selection
   - implicit schedules for initial/naive model exploration
@@ -103,7 +103,7 @@ from finetuning_scheduler import FinetuningScheduler
 trainer = Trainer(callbacks=[FinetuningScheduler()])
 ```
 
-Get started by following [the Finetuning Scheduler introduction](https://finetuning-scheduler.readthedocs.io/en/latest/index.html) which includes a [CLI-based example](https://finetuning-scheduler.readthedocs.io/en/latest/index.html#scheduled-finetuning-superglue) or by following the [notebook-based](https://pytorch-lightning.readthedocs.io/en/latest/notebooks/lightning_examples/finetuning-scheduler.html) Finetuning Scheduler tutorial.
+Get started by following [the Finetuning Scheduler introduction](https://finetuning-scheduler.readthedocs.io/en/stable/index.html) which includes a [CLI-based example](https://finetuning-scheduler.readthedocs.io/en/stable/index.html#scheduled-finetuning-superglue) or by following the [notebook-based](https://pytorch-lightning.readthedocs.io/en/latest/notebooks/lightning_examples/finetuning-scheduler.html) Finetuning Scheduler tutorial.
 
 ______________________________________________________________________
 
@@ -112,7 +112,7 @@ ______________________________________________________________________
 ### Scheduled Finetuning For SuperGLUE
 
 - [Notebook-based Tutorial](https://pytorch-lightning.readthedocs.io/en/latest/notebooks/lightning_examples/finetuning-scheduler.html)
-- [CLI-based Tutorial](https://finetuning-scheduler.readthedocs.io/en/latest/#scheduled-finetuning-superglue)
+- [CLI-based Tutorial](https://finetuning-scheduler.readthedocs.io/en/stable/#scheduled-finetuning-superglue)
 
 ______________________________________________________________________
 
@@ -140,7 +140,7 @@ To ensure maximum stability, the latest PyTorch Lightning patch release fully te
 
 Finetuning Scheduler is developed and maintained by the community in close communication with the [PyTorch Lightning team](https://pytorch-lightning.readthedocs.io/en/latest/governance.html#leads). Thanks to everyone in the community for their tireless effort building and improving the immensely useful core PyTorch Lightning project.
 
-PR's welcome! Please see the [contributing guidelines](https://finetuning-scheduler.readthedocs.io/en/latest/generated/CONTRIBUTING.html) (which are essentially the same as PyTorch Lightning's).
+PR's welcome! Please see the [contributing guidelines](https://finetuning-scheduler.readthedocs.io/en/stable/generated/CONTRIBUTING.html) (which are essentially the same as PyTorch Lightning's).
 
 ______________________________________________________________________
 

docs/source/advanced/lr_scheduler_reinitialization.rst

@@ -113,7 +113,7 @@ sanity-checked prior to training initiation.
 
 Note that specifying LR scheduler reinitialization configurations is only supported for phases >= ``1``. This is because
 for finetuning phase ``0``, the LR scheduler configuration will be the scheduler that you initiate your training session
-with, usually via the ``configure_optimizer`` method of :external+pl:class:`~pytorch_lightning.core.module.LightningModule`.
+with, usually via the ``configure_optimizer`` method of :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule`.
 
 .. tip::
 
@@ -233,7 +233,7 @@ could use:
                 name: Implicit_Reinit_LR_Scheduler
 
 Note that an initial lr scheduler configuration should also still be provided per usual (again, typically via the
-``configure_optimizer`` method of :external+pl:class:`~pytorch_lightning.core.module.LightningModule`) and the initial
+``configure_optimizer`` method of :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule`) and the initial
 lr scheduler configuration can differ in lr scheduler type and configuration from the configuration specified in
 :paramref:`~finetuning_scheduler.fts.FinetuningScheduler.reinit_lr_cfg` applied at each phase transition. Because the
 same schedule is applied at each phase transition, the ``init_pg_lrs`` list is not supported in an implicit finetuning
@@ -277,7 +277,7 @@ training phases:
    = ``1.0e-05``)
 
 Phase ``0`` in :yellow-highlight:`yellow` (passed to our
-:external+pl:class:`~pytorch_lightning.core.module.LightningModule` via the ``model``
+:external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` via the ``model``
 definition in our :external+pl:class:`~pytorch_lightning.utilities.cli.LightningCLI` configuration) uses a
 :external+torch:class:`~torch.optim.lr_scheduler.LinearLR` scheduler (defined in
 ``./config/advanced/fts_explicit_reinit_lr.yaml``) with the initial lr defined via the shared initial optimizer

docs/source/index.rst

@@ -20,7 +20,7 @@ foundational model experimentation with flexible finetuning schedules. Training
    If you're exploring using the :class:`~finetuning_scheduler.fts.FinetuningScheduler`, this is a great place
    to start!
    You may also find the `notebook-based tutorial <https://pytorchlightning.github.io/lightning-tutorials/notebooks/lightning_examples/finetuning-scheduler.html>`_
-   useful and for those using the :doc:`LightningCLI<cli/lightning_cli>`, there is a
+   useful and for those using the :doc:`LightningCLI<common/lightning_cli>`, there is a
    :ref:`CLI-based<scheduled-finetuning-superglue>` example at the bottom of this introduction.
 
 Setup
@@ -92,7 +92,7 @@ thawed/unfrozen parameter groups associated with each finetuning phase as desire
 and executed in ascending order.
 
 1. First, generate the default schedule to ``Trainer.log_dir``. It will be named after your
-   :external+pl:class:`~pytorch_lightning.core.module.LightningModule` subclass with the suffix
+   :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` subclass with the suffix
    ``_ft_schedule.yaml``.
 
 .. code-block:: python
@@ -289,7 +289,7 @@ A demonstration of the scheduled finetuning callback
 :class:`~finetuning_scheduler.fts.FinetuningScheduler` using the
 `RTE <https://huggingface.co/datasets/viewer/?dataset=super_glue&config=rte>`_ and
 `BoolQ <https://github.com/google-research-datasets/boolean-questions>`_ tasks of the
-`SuperGLUE <https://super.gluebenchmark.com/>`_ benchmark and the :doc:`LightningCLI<cli/lightning_cli>`
+`SuperGLUE <https://super.gluebenchmark.com/>`_ benchmark and the :doc:`LightningCLI<common/lightning_cli>`
 is available under ``./fts_examples/``.
 
 Since this CLI-based example requires a few additional packages (e.g. ``transformers``, ``sentencepiece``), you
@@ -446,7 +446,7 @@ Footnotes
    :caption: Examples
 
    Notebook-based Finetuning Scheduler tutorial <https://pytorchlightning.github.io/lightning-tutorials/notebooks/lightning_examples/finetuning-scheduler.html>
-   CLI-based Finetuning Scheduler tutorial <https://finetuning-scheduler.readthedocs.io/en/latest/#example-scheduled-finetuning-for-superglue>
+   CLI-based Finetuning Scheduler tutorial <https://finetuning-scheduler.readthedocs.io/en/stable/#example-scheduled-finetuning-for-superglue>
 
 .. toctree::
    :maxdepth: 1

finetuning_scheduler/__about__.py

@@ -1,7 +1,7 @@
 import time
 
 _this_year = time.strftime("%Y")
-__version__ = "0.1.3"
+__version__ = "0.1.4"
 __author__ = "Dan Dale"
 __author_email__ = "danny.dale@gmail.com"
 __license__ = "Apache-2.0"
@@ -31,7 +31,7 @@
 Documentation
 -------------
 - https://finetuning-scheduler.readthedocs.io/en/stable/
-- https://finetuning-scheduler.readthedocs.io/en/0.1.3/
+- https://finetuning-scheduler.readthedocs.io/en/0.1.4/
 """
 
 __all__ = ["__author__", "__author_email__", "__copyright__", "__docs__", "__homepage__", "__license__", "__version__"]

finetuning_scheduler/fts.py

@@ -92,7 +92,7 @@ def __init__(
                 :ref:`LR Scheduler Reinitialization<explicit-lr-reinitialization-schedule>` for more complex
                 schedule configurations (including per-phase LR scheduler reinitialization). If a schedule is not
                 provided, will generate and execute a default finetuning schedule using the provided
-                :external+pl:class:`~pytorch_lightning.core.module.LightningModule`. See
+                :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule`. See
                 :ref:`the default schedule<index:The Default Finetuning Schedule>`. Defaults to ``None``.
             max_depth: Maximum schedule depth to which the defined finetuning schedule should be executed. Specifying -1
                 or an integer > (number of defined schedule layers) will result in the entire finetuning schedule being
@@ -105,7 +105,7 @@ def __init__(
                 :class:`~finetuning_scheduler.fts_supporters.FTSCheckpoint`) checkpoint
                 before finetuning depth transitions. Defaults to ``True``.
             gen_ft_sched_only: If ``True``, generate the default finetuning schedule to ``Trainer.log_dir`` (it will be
-                named after your :external+pl:class:`~pytorch_lightning.core.module.LightningModule` subclass with
+                named after your :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` subclass with
                 the suffix ``_ft_schedule.yaml``) and exit without training. Typically used to generate a default
                 schedule that will be adjusted by the user before training. Defaults to ``False``.
             epoch_transitions_only: If ``True``, use epoch-driven stopping criteria exclusively (rather than composing
@@ -186,8 +186,8 @@ def freeze_before_training(self, pl_module: "pl.LightningModule") -> None:
         finetuning schedule.
 
         Args:
-            pl_module (:external+pl:class:`~pytorch_lightning.core.module.LightningModule`): The target
-                :external+pl:class:`~pytorch_lightning.core.module.LightningModule` to freeze parameters of
+            pl_module (:external+pl:class:`~pytorch_lightning.core.lightning.LightningModule`): The target
+                :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` to freeze parameters of
         """
         self.freeze(modules=pl_module)
 
@@ -308,15 +308,15 @@ def setup(self, trainer: "pl.Trainer", pl_module: "pl.LightningModule", stage: O
         1. configure the :class:`~finetuning_scheduler.fts_supporters.FTSEarlyStopping`
         callback (if relevant)
         2. initialize the :attr:`~finetuning_scheduler.fts.FinetuningScheduler._fts_state`
-        3. freeze the target :external+pl:class:`~pytorch_lightning.core.module.LightningModule` parameters
+        3. freeze the target :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` parameters
         Finally, initialize the :class:`~finetuning_scheduler.fts.FinetuningScheduler`
         training session in the training environment.
 
         Args:
             trainer (:external+pl:class:`~pytorch_lightning.trainer.trainer.Trainer`): The
                 :external+pl:class:`~pytorch_lightning.trainer.trainer.Trainer` object
-            pl_module (:external+pl:class:`~pytorch_lightning.core.module.LightningModule`): The
-                :external+pl:class:`~pytorch_lightning.core.module.LightningModule` object
+            pl_module (:external+pl:class:`~pytorch_lightning.core.lightning.LightningModule`): The
+                :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` object
             stage: The ``RunningStage.{SANITY_CHECKING,TRAINING,VALIDATING}``. Defaults to None.
 
         Raises:
@@ -366,8 +366,8 @@ def on_fit_start(self, trainer: "pl.Trainer", pl_module: "pl.LightningModule") -
         Args:
             trainer (:external+pl:class:`~pytorch_lightning.trainer.trainer.Trainer`): The
                 :external+pl:class:`~pytorch_lightning.trainer.trainer.Trainer` object
-            pl_module (:external+pl:class:`~pytorch_lightning.core.module.LightningModule`): The
-                :external+pl:class:`~pytorch_lightning.core.module.LightningModule` object
+            pl_module (:external+pl:class:`~pytorch_lightning.core.lightning.LightningModule`): The
+                :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` object
 
         Raises:
             MisconfigurationException: If more than 1 optimizers are configured indicates a configuration error
@@ -461,8 +461,8 @@ def on_train_epoch_start(self, trainer: "pl.Trainer", pl_module: "pl.LightningMo
         Args:
             trainer (:external+pl:class:`~pytorch_lightning.trainer.trainer.Trainer`): The
                 :external+pl:class:`~pytorch_lightning.trainer.trainer.Trainer` object
-            pl_module (:external+pl:class:`~pytorch_lightning.core.module.LightningModule`): The
-                :external+pl:class:`~pytorch_lightning.core.module.LightningModule` object
+            pl_module (:external+pl:class:`~pytorch_lightning.core.lightning.LightningModule`): The
+                :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` object
         """
         # if resuming from a ckpt, we need to sync fts_state
         if self._fts_state._resume_fit_from_ckpt:
@@ -503,8 +503,8 @@ def on_before_zero_grad(self, trainer: "pl.Trainer", pl_module: "pl.LightningMod
         Args:
             trainer (:external+pl:class:`~pytorch_lightning.trainer.trainer.Trainer`): The
                 :external+pl:class:`~pytorch_lightning.trainer.trainer.Trainer` object
-            pl_module  (:external+pl:class:`~pytorch_lightning.core.module.LightningModule`): The
-                :external+pl:class:`~pytorch_lightning.core.module.LightningModule` object
+            pl_module  (:external+pl:class:`~pytorch_lightning.core.lightning.LightningModule`): The
+                :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` object
             optimizer (:class:`~torch.optim.Optimizer`): The :class:`~torch.optim.Optimizer` to which parameter groups
                 will be configured and added.
         """
@@ -516,7 +516,7 @@ def on_train_end(self, trainer: "pl.Trainer", pl_module: "pl.LightningModule") -
 
         Args:
             trainer (:external+pl:class:`~pytorch_lightning.trainer.trainer.Trainer`): _description_
-            pl_module (:external+pl:class:`~pytorch_lightning.core.module.LightningModule`): _description_
+            pl_module (:external+pl:class:`~pytorch_lightning.core.lightning.LightningModule`): _description_
         """
         assert self._fts_state._ft_sync_objects is not None
         self.sync(self._fts_state._ft_sync_objects, self._fts_state._ft_sync_props)

finetuning_scheduler/fts_supporters.py

@@ -944,7 +944,7 @@ def save_schedule(schedule_name: str, layer_config: Dict, dump_loc: Union[str, o
 
         Returns:
             os.PathLike: The path to the generated schedule, by default ``Trainer.log_dir`` and named after the
-            :external+pl:class:`~pytorch_lightning.core.module.LightningModule` subclass in use with the suffix
+            :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` subclass in use with the suffix
             ``_ft_schedule.yaml``)
         """
         dump_path = pathlib.Path(dump_loc)
@@ -967,7 +967,7 @@ def gen_ft_schedule(module: Module, dump_loc: Union[str, os.PathLike]) -> os.Pat
             dump_loc: The directory to which the generated schedule (.yaml) should be written
         Returns:
             os.PathLike: The path to the generated schedule, by default ``Trainer.log_dir`` and named after the
-            :external+pl:class:`~pytorch_lightning.core.module.LightningModule` subclass in use with the suffix
+            :external+pl:class:`~pytorch_lightning.core.lightning.LightningModule` subclass in use with the suffix
             ``_ft_schedule.yaml``)
         """
         # Note: This initial default finetuning schedule generation approach is intentionally simple/naive but is