pytorch-optimizer 3.3.3__tar.gz → 3.4.0__tar.gz

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/PKG-INFO

@@ -1,9 +1,9 @@
 Metadata-Version: 2.3
 Name: pytorch_optimizer
-Version: 3.3.3
+Version: 3.4.0
 Summary: optimizer & lr scheduler & objective function collections in PyTorch
 License: Apache-2.0
-Keywords: pytorch,deep-learning,optimizer,lr scheduler,A2Grad,ASGD,AccSGD,AdaBelief,AdaBound,AdaDelta,AdaFactor,AdaMax,AdamG,AdaMod,AdaNorm,AdaPNM,AdaSmooth,AdEMAMix,ADOPT,AdaHessian,Adai,Adalite,AdaLomo,AdamMini,AdamP,AdamS,Adan,AggMo,Aida,AliG,Amos,Apollo,APOLLO,AvaGrad,bSAM,CAME,DAdaptAdaGrad,DAdaptAdam,DAdaptAdan,DAdaptSGD,DAdaptLion,DeMo,DiffGrad,FAdam,Fromage,FTRL,GaLore,Grams,Gravity,GrokFast,GSAM,Kate,Lamb,LaProp,LARS,Lion,LOMO,Lookahead,MADGRAD,MARS,MSVAG,Muno,Nero,NovoGrad,OrthoGrad,PAdam,PCGrad,PID,PNM,Prodigy,QHAdam,QHM,RAdam,Ranger,Ranger21,RotoGrad,SAM,ScheduleFreeSGD,ScheduleFreeAdamW,ScheduleFreeRAdam,SGDP,Shampoo,ScalableShampoo,SGDW,SignSGD,SM3,SOAP,SopihaH,SRMM,StableAdamW,SWATS,Tiger,TRAC,WSAM,Yogi,BCE,BCEFocal,Focal,FocalCosine,SoftF1,Dice,LDAM,Jaccard,Bi-Tempered,Tversky,FocalTversky,LovaszHinge,bitsandbytes,WSD,QGaLore
+Keywords: pytorch,deep-learning,optimizer,lr scheduler,A2Grad,ASGD,AccSGD,AdaBelief,AdaBound,AdaDelta,AdaFactor,AdaMax,AdamG,AdaMod,AdaNorm,AdaPNM,AdaSmooth,AdEMAMix,ADOPT,AdaHessian,Adai,Adalite,AdaLomo,AdamMini,AdamP,AdamS,Adan,AggMo,Aida,AliG,Amos,Apollo,APOLLO,AvaGrad,bSAM,CAME,DAdaptAdaGrad,DAdaptAdam,DAdaptAdan,DAdaptSGD,DAdaptLion,DeMo,DiffGrad,EXAdam,FAdam,FOCUS,Fromage,FTRL,GaLore,Grams,Gravity,GrokFast,GSAM,Kate,Lamb,LaProp,LARS,Lion,LOMO,Lookahead,MADGRAD,MARS,MSVAG,Muno,Nero,NovoGrad,OrthoGrad,PAdam,PCGrad,PID,PNM,Prodigy,PSGD,QHAdam,QHM,RAdam,Ranger,Ranger21,RotoGrad,SAM,ScheduleFreeSGD,ScheduleFreeAdamW,ScheduleFreeRAdam,SGDP,Shampoo,ScalableShampoo,SGDW,SignSGD,SM3,SOAP,SopihaH,SPAM,SRMM,StableAdamW,SWATS,TAM,Tiger,TRAC,WSAM,Yogi,BCE,BCEFocal,Focal,FocalCosine,SoftF1,Dice,LDAM,Jaccard,Bi-Tempered,Tversky,FocalTversky,LovaszHinge,bitsandbytes,WSD,QGaLore
 Author: kozistr
 Author-email: kozistr@gmail.com
 Maintainer: kozistr
@@ -29,8 +29,6 @@ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Classifier: Topic :: Software Development
 Classifier: Topic :: Software Development :: Libraries
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
-Provides-Extra: bitsandbytes
-Requires-Dist: bitsandbytes (>=0.44,<0.45) ; extra == "bitsandbytes"
 Requires-Dist: numpy (<=1.24.4) ; python_version < "3.9"
 Requires-Dist: numpy (>1.24.4) ; python_version >= "3.9"
 Requires-Dist: torch (>=1.10) ; python_version >= "3.8"
@@ -51,8 +49,8 @@ Description-Content-Type: text/markdown
 
 ## The reasons why you use `pytorch-optimizer`.
 
-* Wide range of supported optimizers. Currently, **89 optimizers (+ `bitsandbytes`, `qgalore`, `torchao`)**, **16 lr schedulers**, and **13 loss functions** are supported!
-* Including many variants such as `Cautious`, `AdamD`, `Gradient Centrailiaztion`
+* Wide range of supported optimizers. Currently, **95 optimizers (+ `bitsandbytes`, `qgalore`, `torchao`)**, **16 lr schedulers**, and **13 loss functions** are supported!
+* Including many variants such as `ADOPT`, `Cautious`, `AdamD`, `StableAdamW`, and `Gradient Centrailiaztion`
 * Easy to use, clean, and tested codes
 * Active maintenance
 * Somewhat a bit more optimized compared to the original implementation
@@ -239,6 +237,11 @@ get_supported_optimizers(['adam*', 'ranger*'])
 | Grams         | *Gradient Descent with Adaptive Momentum Scaling*                                                 |                                                                                                                | <https://arxiv.org/abs/2412.17107>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2024arXiv241217107C/exportcitation)                                                        |
 | OrthoGrad     | *Grokking at the Edge of Numerical Stability*                                                     | [github](https://github.com/LucasPrietoAl/grokking-at-the-edge-of-numerical-stability)                         | <https://arxiv.org/abs/2501.04697>                                                          | [cite](https://github.com/LucasPrietoAl/grokking-at-the-edge-of-numerical-stability?tab=readme-ov-file#citation)                    |
 | Adam-ATAN2    | *Scaling Exponents Across Parameterizations and Optimizers*                                       |                                                                                                                | <https://arxiv.org/abs/2407.05872>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2024arXiv240705872E/exportcitation)                                                        |
+| SPAM          | *Spike-Aware Adam with Momentum Reset for Stable LLM Training*                                    | [github](https://github.com/TianjinYellow/SPAM-Optimizer)                                                      | <https://arxiv.org/abs/2501.06842>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2025arXiv250106842H/exportcitation)                                                        |
+| TAM           | *Torque-Aware Momentum*                                                                           |                                                                                                                | <https://arxiv.org/abs/2412.18790>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2024arXiv241218790M/exportcitation)                                                        |
+| FOCUS         | *First Order Concentrated Updating Scheme*                                                        | [github](https://github.com/liuyz0/FOCUS)                                                                      | <https://arxiv.org/abs/2501.12243>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2025arXiv250112243M/exportcitation)                                                        |
+| PSGD          | *Preconditioned Stochastic Gradient Descent*                                                      | [github](https://github.com/lixilinx/psgd_torch)                                                               | <https://arxiv.org/abs/1512.04202>                                                          | [cite](https://github.com/lixilinx/psgd_torch?tab=readme-ov-file#resources)                                                         |
+| EXAdam        | *The Power of Adaptive Cross-Moments*                                                             | [github](https://github.com/AhmedMostafa16/EXAdam)                                                             | <https://arxiv.org/abs/2412.20302>                                                          | [cite](https://github.com/AhmedMostafa16/EXAdam?tab=readme-ov-file#citation)                                                        |
 
 ## Supported LR Scheduler
 

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/README.md

@@ -10,8 +10,8 @@
 
 ## The reasons why you use `pytorch-optimizer`.
 
-* Wide range of supported optimizers. Currently, **89 optimizers (+ `bitsandbytes`, `qgalore`, `torchao`)**, **16 lr schedulers**, and **13 loss functions** are supported!
-* Including many variants such as `Cautious`, `AdamD`, `Gradient Centrailiaztion`
+* Wide range of supported optimizers. Currently, **95 optimizers (+ `bitsandbytes`, `qgalore`, `torchao`)**, **16 lr schedulers**, and **13 loss functions** are supported!
+* Including many variants such as `ADOPT`, `Cautious`, `AdamD`, `StableAdamW`, and `Gradient Centrailiaztion`
 * Easy to use, clean, and tested codes
 * Active maintenance
 * Somewhat a bit more optimized compared to the original implementation
@@ -198,6 +198,11 @@ get_supported_optimizers(['adam*', 'ranger*'])
 | Grams         | *Gradient Descent with Adaptive Momentum Scaling*                                                 |                                                                                                                | <https://arxiv.org/abs/2412.17107>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2024arXiv241217107C/exportcitation)                                                        |
 | OrthoGrad     | *Grokking at the Edge of Numerical Stability*                                                     | [github](https://github.com/LucasPrietoAl/grokking-at-the-edge-of-numerical-stability)                         | <https://arxiv.org/abs/2501.04697>                                                          | [cite](https://github.com/LucasPrietoAl/grokking-at-the-edge-of-numerical-stability?tab=readme-ov-file#citation)                    |
 | Adam-ATAN2    | *Scaling Exponents Across Parameterizations and Optimizers*                                       |                                                                                                                | <https://arxiv.org/abs/2407.05872>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2024arXiv240705872E/exportcitation)                                                        |
+| SPAM          | *Spike-Aware Adam with Momentum Reset for Stable LLM Training*                                    | [github](https://github.com/TianjinYellow/SPAM-Optimizer)                                                      | <https://arxiv.org/abs/2501.06842>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2025arXiv250106842H/exportcitation)                                                        |
+| TAM           | *Torque-Aware Momentum*                                                                           |                                                                                                                | <https://arxiv.org/abs/2412.18790>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2024arXiv241218790M/exportcitation)                                                        |
+| FOCUS         | *First Order Concentrated Updating Scheme*                                                        | [github](https://github.com/liuyz0/FOCUS)                                                                      | <https://arxiv.org/abs/2501.12243>                                                          | [cite](https://ui.adsabs.harvard.edu/abs/2025arXiv250112243M/exportcitation)                                                        |
+| PSGD          | *Preconditioned Stochastic Gradient Descent*                                                      | [github](https://github.com/lixilinx/psgd_torch)                                                               | <https://arxiv.org/abs/1512.04202>                                                          | [cite](https://github.com/lixilinx/psgd_torch?tab=readme-ov-file#resources)                                                         |
+| EXAdam        | *The Power of Adaptive Cross-Moments*                                                             | [github](https://github.com/AhmedMostafa16/EXAdam)                                                             | <https://arxiv.org/abs/2412.20302>                                                          | [cite](https://github.com/AhmedMostafa16/EXAdam?tab=readme-ov-file#citation)                                                        |
 
 ## Supported LR Scheduler
 

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pytorch_optimizer"
-version = "3.3.3"
+version = "3.4.0"
 description = "optimizer & lr scheduler & objective function collections in PyTorch"
 license = "Apache-2.0"
 authors = ["kozistr <kozistr@gmail.com>"]
@@ -14,13 +14,13 @@ keywords = [
     "AdaDelta", "AdaFactor", "AdaMax", "AdamG", "AdaMod", "AdaNorm", "AdaPNM", "AdaSmooth", "AdEMAMix", "ADOPT",
     "AdaHessian", "Adai", "Adalite", "AdaLomo", "AdamMini", "AdamP", "AdamS", "Adan", "AggMo", "Aida", "AliG", "Amos",
     "Apollo", "APOLLO", "AvaGrad", "bSAM", "CAME", "DAdaptAdaGrad", "DAdaptAdam", "DAdaptAdan", "DAdaptSGD",
-    "DAdaptLion", "DeMo", "DiffGrad", "FAdam", "Fromage", "FTRL", "GaLore", "Grams", "Gravity", "GrokFast", "GSAM",
-    "Kate", "Lamb", "LaProp", "LARS", "Lion", "LOMO", "Lookahead", "MADGRAD", "MARS", "MSVAG", "Muno", "Nero",
-    "NovoGrad", "OrthoGrad", "PAdam", "PCGrad", "PID", "PNM", "Prodigy", "QHAdam", "QHM", "RAdam", "Ranger",
-    "Ranger21", "RotoGrad", "SAM", "ScheduleFreeSGD", "ScheduleFreeAdamW", "ScheduleFreeRAdam", "SGDP", "Shampoo",
-    "ScalableShampoo", "SGDW", "SignSGD", "SM3", "SOAP", "SopihaH", "SRMM", "StableAdamW", "SWATS", "Tiger", "TRAC",
-    "WSAM", "Yogi", "BCE", "BCEFocal", "Focal", "FocalCosine", "SoftF1", "Dice", "LDAM", "Jaccard", "Bi-Tempered",
-    "Tversky", "FocalTversky", "LovaszHinge", "bitsandbytes", "WSD", "QGaLore",
+    "DAdaptLion", "DeMo", "DiffGrad", "EXAdam", "FAdam", "FOCUS", "Fromage", "FTRL", "GaLore", "Grams", "Gravity",
+    "GrokFast", "GSAM", "Kate", "Lamb", "LaProp", "LARS", "Lion", "LOMO", "Lookahead", "MADGRAD", "MARS", "MSVAG",
+    "Muno", "Nero", "NovoGrad", "OrthoGrad", "PAdam", "PCGrad", "PID", "PNM", "Prodigy", "PSGD", "QHAdam", "QHM",
+    "RAdam", "Ranger", "Ranger21", "RotoGrad", "SAM", "ScheduleFreeSGD", "ScheduleFreeAdamW", "ScheduleFreeRAdam",
+    "SGDP", "Shampoo", "ScalableShampoo", "SGDW", "SignSGD", "SM3", "SOAP", "SopihaH", "SPAM", "SRMM", "StableAdamW",
+    "SWATS", "TAM", "Tiger", "TRAC", "WSAM", "Yogi", "BCE", "BCEFocal", "Focal", "FocalCosine", "SoftF1", "Dice",
+    "LDAM", "Jaccard", "Bi-Tempered", "Tversky", "FocalTversky", "LovaszHinge", "bitsandbytes", "WSD", "QGaLore",
 ]
 classifiers = [
     "License :: OSI Approved :: Apache Software License",
@@ -52,29 +52,27 @@ numpy = [
     { version = "<=1.24.4", python = "<3.9" },
 ]
 torch = { version = ">=1.10", python = ">=3.8", source = "torch" }
-bitsandbytes = { version = "^0.44", optional = true }
 
-[tool.poetry.dev-dependencies]
-isort = { version = "^5", python = ">=3.8" }
-black = { version = "^24", python = ">=3.8" }
+[tool.poetry.group.dev.dependencies]
+isort = [
+    { version = "<6", python = "<3.9" },
+    { version = "^6", python = ">=3.9" },
+]
+black = [
+    { version = "<25", python = "<3.9" },
+    { version = "^25", python = ">=3.9" },
+]
 ruff = "*"
 pytest = "*"
 pytest-cov = "*"
 
-[tool.poetry.extras]
-bitsandbytes = ["bitsandbytes"]
-
 [[tool.poetry.source]]
 name = "torch"
 url = "https://download.pytorch.org/whl/cpu"
 priority = "explicit"
 
 [tool.ruff]
-src = [
-    "pytorch_optimizer",
-    "tests",
-    "examples",
-]
+src = ["pytorch_optimizer", "tests", "examples"]
 target-version = "py312"
 line-length = 119
 exclude = [
@@ -97,8 +95,10 @@ select = [
     "TID", "ARG", "ERA", "RUF", "YTT", "PL", "Q"
 ]
 ignore = [
-    "A005", "B905", "D100", "D102", "D104", "D105", "D107", "D203", "D213", "D413", "PIE790", "PLR0912", "PLR0913",
-    "PLR0915", "PLR2004", "RUF013", "Q003", "ARG002",
+    "A005", "B905",
+    "D100", "D102", "D104", "D105", "D107", "D203", "D213", "D413",
+    "PLR0912", "PLR0913", "PLR0915", "PLR2004",
+    "Q003", "ARG002",
 ]
 fixable = ["ALL"]
 unfixable = ["F401"]
@@ -107,10 +107,10 @@ flake8-quotes.docstring-quotes = "double"
 flake8-quotes.inline-quotes = "single"
 
 [tool.ruff.lint.extend-per-file-ignores]
-"hubconf.py" = ["D", "INP001"]
 "examples/visualize_optimizers.py" = ["D103", "D400", "D415"]
-"**/__init__.py" = ["F401"]
 "{tests,examples}/*.py" = ["D", "S101"]
+"**/__init__.py" = ["F401"]
+"hubconf.py" = ["D", "INP001"]
 
 [tool.ruff.lint.isort]
 combine-as-imports = false
@@ -129,9 +129,9 @@ testpaths = "tests"
 
 [tool.coverage.run]
 omit = [
-    "./pytorch_optimizer/optimizer/rotograd.py",
     "./pytorch_optimizer/optimizer/adam_mini.py",
     "./pytorch_optimizer/optimizer/demo.py",
+    "./pytorch_optimizer/optimizer/rotograd.py",
 ]
 
 [build-system]

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/__init__.py

@@ -46,6 +46,7 @@ from pytorch_optimizer.optimizer import (
     ASGD,
     BSAM,
     CAME,
+    FOCUS,
     FTRL,
     GSAM,
     LARS,
@@ -61,8 +62,10 @@ from pytorch_optimizer.optimizer import (
     SGDW,
     SM3,
     SOAP,
+    SPAM,
     SRMM,
     SWATS,
+    TAM,
     TRAC,
     WSAM,
     A2Grad,
@@ -87,6 +90,7 @@ from pytorch_optimizer.optimizer import (
     AdaPNM,
     AdaShift,
     AdaSmooth,
+    AdaTAM,
     AdEMAMix,
     AggMo,
     Aida,
@@ -102,6 +106,7 @@ from pytorch_optimizer.optimizer import (
     DeMo,
     DiffGrad,
     DynamicLossScaler,
+    EXAdam,
     FAdam,
     Fromage,
     GaLore,
@@ -109,6 +114,7 @@ from pytorch_optimizer.optimizer import (
     Gravity,
     GrokFastAdamW,
     Kate,
+    Kron,
     Lamb,
     LaProp,
     Lion,

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/base/optimizer.py

@@ -6,7 +6,16 @@ import torch
 from torch.optim import Optimizer
 
 from pytorch_optimizer.base.exception import NegativeLRError, NegativeStepError
-from pytorch_optimizer.base.types import BETAS, CLOSURE, DEFAULTS, HUTCHINSON_G, LOSS, PARAMETERS, STATE
+from pytorch_optimizer.base.types import (
+    BETAS,
+    CLOSURE,
+    DEFAULTS,
+    HUTCHINSON_G,
+    LOSS,
+    OPTIMIZER_INSTANCE_OR_CLASS,
+    PARAMETERS,
+    STATE,
+)
 
 
 class BaseOptimizer(ABC, Optimizer):
@@ -15,6 +24,18 @@ class BaseOptimizer(ABC, Optimizer):
     def __init__(self, params: PARAMETERS, defaults: DEFAULTS) -> None:
         super().__init__(params, defaults)
 
+    @staticmethod
+    def load_optimizer(optimizer: OPTIMIZER_INSTANCE_OR_CLASS, **kwargs) -> Optimizer:
+        r"""Build torch.optim.Optimizer class."""
+        if isinstance(optimizer, Optimizer):
+            return optimizer
+
+        if 'params' in kwargs:
+            params = kwargs.pop('params')
+            return optimizer(params, **kwargs)
+
+        raise ValueError('need to pass `params` when you pass the `torch.optim.Optimizer` instance.')
+
     @staticmethod
     @torch.no_grad()
     def set_hessian(param_groups: PARAMETERS, state: STATE, hessian: List[torch.Tensor]) -> None:
@@ -317,7 +338,7 @@ class BaseOptimizer(ABC, Optimizer):
     @staticmethod
     def validate_options(x: str, name: str, options: List[str]) -> None:
         if x not in options:
-            opts: str = ' or '.join([f'\'{option}\'' for option in options]).strip()
+            opts: str = ' or '.join([f"'{option}'" for option in options]).strip()
             raise ValueError(f'[-] {name} {x} must be one of ({opts})')
 
     @staticmethod

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/base/scheduler.py

@@ -1,8 +1,9 @@
 from abc import ABC, abstractmethod
 from typing import List
 
+from torch.optim import Optimizer
+
 from pytorch_optimizer.base.exception import NegativeLRError, NegativeStepError
-from pytorch_optimizer.base.types import OPTIMIZER
 
 
 class BaseLinearWarmupScheduler(ABC):
@@ -10,7 +11,7 @@ class BaseLinearWarmupScheduler(ABC):
 
         The LR Scheduler class based on this class has linear warmup strategy.
 
-    :param optimizer: Optimizer. OPTIMIZER. It will set learning rate to all trainable parameters in optimizer.
+    :param optimizer: Optimizer. It will set learning rate to all trainable parameters in optimizer.
     :param t_max: int. total steps to train.
     :param max_lr: float. maximum lr.
     :param min_lr: float. minimum lr.
@@ -20,7 +21,7 @@ class BaseLinearWarmupScheduler(ABC):
 
     def __init__(
         self,
-        optimizer: OPTIMIZER,
+        optimizer: Optimizer,
         t_max: int,
         max_lr: float,
         min_lr: float = 0.0,

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/base/types.py

@@ -11,6 +11,7 @@ DEFAULTS = Dict
 PARAMETERS = Optional[Union[Iterable[Dict], Iterable[torch.Tensor]]]
 STATE = Dict
 OPTIMIZER = Type[Optimizer]
+OPTIMIZER_INSTANCE_OR_CLASS = Union[OPTIMIZER, Optimizer]
 SCHEDULER = Type[LRScheduler]
 
 HUTCHINSON_G = Literal['gaussian', 'rademacher']

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/loss/jaccard.py

@@ -49,7 +49,7 @@ class JaccardLoss(_Loss):
     def __init__(
         self,
         mode: CLASS_MODE,
-        classes: List[int] = None,
+        classes: Optional[List[int]] = None,
         log_loss: bool = False,
         from_logits: bool = True,
         label_smooth: float = 0.0,
@@ -59,7 +59,7 @@ class JaccardLoss(_Loss):
 
         if classes is not None:
             if mode == 'binary':
-                raise ValueError('[-] Masking classes is not supported with mode=binary')
+                raise ValueError('masking classes is not supported with mode=binary')
 
             classes = torch.LongTensor(classes)
 

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/lr_scheduler/cosine_anealing.py

@@ -1,10 +1,9 @@
 import math
 from typing import List, Optional
 
+from torch.optim import Optimizer
 from torch.optim.lr_scheduler import LRScheduler
 
-from pytorch_optimizer.base.types import OPTIMIZER
-
 
 class CosineAnnealingWarmupRestarts(LRScheduler):
     r"""CosineAnnealingWarmupRestarts.
@@ -21,7 +20,7 @@ class CosineAnnealingWarmupRestarts(LRScheduler):
 
     def __init__(
         self,
-        optimizer: OPTIMIZER,
+        optimizer: Optimizer,
         first_cycle_steps: int,
         cycle_mult: float = 1.0,
         max_lr: float = 1e-4,
@@ -53,7 +52,7 @@ class CosineAnnealingWarmupRestarts(LRScheduler):
 
         self.init_lr()
 
-    def init_lr(self):
+    def init_lr(self) -> None:
         self.base_lrs = []
         for param_group in self.optimizer.param_groups:
             param_group['lr'] = self.min_lr

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/lr_scheduler/rex.py

@@ -1,9 +1,8 @@
 from typing import List, Optional
 
+from torch.optim import Optimizer
 from torch.optim.lr_scheduler import LRScheduler
 
-from pytorch_optimizer.base.types import OPTIMIZER
-
 
 class REXScheduler(LRScheduler):
     r"""Revisiting Budgeted Training with an Improved Schedule.
@@ -16,7 +15,7 @@ class REXScheduler(LRScheduler):
 
     def __init__(
         self,
-        optimizer: OPTIMIZER,
+        optimizer: Optimizer,
         total_steps: int,
         max_lr: float = 1.0,
         min_lr: float = 0.0,
@@ -35,7 +34,7 @@ class REXScheduler(LRScheduler):
 
         self.init_lr()
 
-    def init_lr(self):
+    def init_lr(self) -> None:
         self.base_lrs = []
         for param_group in self.optimizer.param_groups:
             param_group['lr'] = self.min_lr

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/optimizer/__init__.py

@@ -1,10 +1,10 @@
 import fnmatch
 from importlib.util import find_spec
-from typing import Dict, List, Optional, Sequence, Set, Union
+from typing import Dict, List, Optional, Sequence, Set, Type, Union
 
 import torch
 from torch import nn
-from torch.optim import AdamW
+from torch.optim import AdamW, Optimizer
 
 from pytorch_optimizer.base.types import OPTIMIZER, PARAMETERS
 from pytorch_optimizer.optimizer.a2grad import A2Grad
@@ -40,8 +40,10 @@ from pytorch_optimizer.optimizer.came import CAME
 from pytorch_optimizer.optimizer.dadapt import DAdaptAdaGrad, DAdaptAdam, DAdaptAdan, DAdaptLion, DAdaptSGD
 from pytorch_optimizer.optimizer.demo import DeMo
 from pytorch_optimizer.optimizer.diffgrad import DiffGrad
+from pytorch_optimizer.optimizer.exadam import EXAdam
 from pytorch_optimizer.optimizer.experimental.ranger25 import Ranger25
 from pytorch_optimizer.optimizer.fadam import FAdam
+from pytorch_optimizer.optimizer.focus import FOCUS
 from pytorch_optimizer.optimizer.fp16 import DynamicLossScaler, SafeFP16Optimizer
 from pytorch_optimizer.optimizer.fromage import Fromage
 from pytorch_optimizer.optimizer.ftrl import FTRL
@@ -69,6 +71,7 @@ from pytorch_optimizer.optimizer.pcgrad import PCGrad
 from pytorch_optimizer.optimizer.pid import PID
 from pytorch_optimizer.optimizer.pnm import PNM
 from pytorch_optimizer.optimizer.prodigy import Prodigy
+from pytorch_optimizer.optimizer.psgd import Kron
 from pytorch_optimizer.optimizer.qhadam import QHAdam
 from pytorch_optimizer.optimizer.qhm import QHM
 from pytorch_optimizer.optimizer.radam import RAdam
@@ -83,8 +86,10 @@ from pytorch_optimizer.optimizer.shampoo import ScalableShampoo, Shampoo
 from pytorch_optimizer.optimizer.sm3 import SM3
 from pytorch_optimizer.optimizer.soap import SOAP
 from pytorch_optimizer.optimizer.sophia import SophiaH
+from pytorch_optimizer.optimizer.spam import SPAM
 from pytorch_optimizer.optimizer.srmm import SRMM
 from pytorch_optimizer.optimizer.swats import SWATS
+from pytorch_optimizer.optimizer.tam import TAM, AdaTAM
 from pytorch_optimizer.optimizer.tiger import Tiger
 from pytorch_optimizer.optimizer.trac import TRAC
 from pytorch_optimizer.optimizer.yogi import Yogi
@@ -251,6 +256,8 @@ OPTIMIZER_LIST: List[OPTIMIZER] = [
     SRMM,
     AvaGrad,
     AdaShift,
+    TAM,
+    AdaTAM,
     AdaDelta,
     Amos,
     AdaHessian,
@@ -285,7 +292,11 @@ OPTIMIZER_LIST: List[OPTIMIZER] = [
     LaProp,
     MARS,
     SGDSaI,
+    FOCUS,
     Grams,
+    SPAM,
+    Kron,
+    EXAdam,
     Ranger25,
 ]
 OPTIMIZERS: Dict[str, OPTIMIZER] = {str(optimizer.__name__).lower(): optimizer for optimizer in OPTIMIZER_LIST}
@@ -298,8 +309,9 @@ def create_optimizer(
     weight_decay: float = 0.0,
     wd_ban_list: List[str] = ('bias', 'LayerNorm.bias', 'LayerNorm.weight'),
     use_lookahead: bool = False,
+    use_orthograd: bool = False,
     **kwargs,
-):
+) -> Optimizer:
     r"""Build optimizer.
 
     :param model: nn.Module. model.
@@ -307,7 +319,8 @@ def create_optimizer(
     :param lr: float. learning rate.
     :param weight_decay: float. weight decay.
     :param wd_ban_list: List[str]. weight decay ban list by layer.
-    :param use_lookahead: bool. use lookahead.
+    :param use_lookahead: bool. use Lookahead.
+    :param use_orthograd: bool. use OrthoGrad.
     """
     optimizer_name = optimizer_name.lower()
 
@@ -315,14 +328,17 @@ def create_optimizer(
         get_optimizer_parameters(model, weight_decay, wd_ban_list) if weight_decay > 0.0 else model.parameters()
     )
 
-    optimizer = load_optimizer(optimizer_name)
+    optimizer_class: OPTIMIZER = load_optimizer(optimizer_name)
 
     if optimizer_name == 'alig':
-        optimizer = optimizer(parameters, max_lr=lr, **kwargs)
+        optimizer = optimizer_class(parameters, max_lr=lr, **kwargs)
     elif optimizer_name in {'lomo', 'adalomo', 'adammini'}:
-        optimizer = optimizer(model, lr=lr, **kwargs)
+        optimizer = optimizer_class(model, lr=lr, **kwargs)
     else:
-        optimizer = optimizer(parameters, lr=lr, **kwargs)
+        optimizer = optimizer_class(parameters, lr=lr, **kwargs)
+
+    if use_orthograd:
+        optimizer = OrthoGrad(optimizer, **kwargs)
 
     if use_lookahead:
         optimizer = Lookahead(

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/optimizer/adafactor.py

@@ -13,8 +13,8 @@ class AdaFactor(BaseOptimizer):
 
     :param params: PARAMETERS. iterable of parameters to optimize or dicts defining parameter groups.
     :param lr: float. learning rate.
-    :param betas: BETAS. coefficients used for computing running averages of gradient and the squared
-        hessian trace. if beta1 is None, first momentum will be skipped.
+    :param betas: BETAS. coefficients used for computing running averages of gradient and the squared hessian trace.
+        if beta1 is None, first momentum will be skipped.
     :param decay_rate: float. coefficient used to compute running averages of square gradient.
     :param weight_decay: float. weight decay (L2 penalty).
     :param weight_decouple: bool. the optimizer uses decoupled weight decay as in AdamW.

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/optimizer/adamp.py

@@ -141,6 +141,10 @@ class AdamP(BaseOptimizer):
                 inv_de_nom = exp_avg_sq.rsqrt().add_(group['eps']).mul_(bias_correction2_sq)
 
                 perturb = exp_avg.clone()
+
+                if self.cautious:
+                    self.apply_cautious(perturb, grad)
+
                 if group['nesterov']:
                     perturb.mul_(beta1).addcmul_(grad, inv_de_nom, value=1.0 - beta1)
                 else:
@@ -173,9 +177,6 @@ class AdamP(BaseOptimizer):
                     bias_correction1=bias_correction1,
                 )
 
-                if self.cautious:
-                    self.apply_cautious(perturb, grad)
-
                 p.add_(perturb, alpha=-step_size)
 
         return loss

{pytorch_optimizer-3.3.3 → pytorch_optimizer-3.4.0}/pytorch_optimizer/optimizer/adamw.py

@@ -107,8 +107,7 @@ class StableAdamW(BaseOptimizer):
                 exp_avg.lerp_(grad, weight=beta1_comp)
                 exp_avg_sq.mul_(beta2_hat).addcmul_(grad, grad, value=1.0 - beta2_hat)
 
-                rms = self.get_stable_adamw_rms(grad, exp_avg_sq, eps=eps_p2)
-                lr = group['lr'] / rms
+                lr: float = group['lr'] / self.get_stable_adamw_rms(grad, exp_avg_sq, eps=eps_p2)
 
                 self.apply_weight_decay(
                     p,