An impl of adafactor as per big vision (scaling vit) changes

Author: rwightman

timm/optim/__init__.py

@@ -4,6 +4,7 @@
 from .adamp import AdamP
 from .adamw import AdamW
 from .adan import Adan
+from .adafactor_bv import AdafactorBigVision
 from .lamb import Lamb
 from .lars import Lars
 from .lookahead import Lookahead

timm/optim/adafactor_bv.py

@@ -0,0 +1,288 @@
+from typing import List, Optional, Tuple, Union
+
+import numpy as np
+import torch
+from torch import Tensor
+from torch.optim import Optimizer
+
+
+def _get_scalar_dtype():
+    """Get the scalar dtype that the optimizer uses for state"""
+    return torch.float64
+
+def _factored_dims(
+    shape: Tuple[int, ...],
+    factored: bool,
+    min_dim_size_to_factor: int
+) -> Optional[tuple[int, int]]:
+  """Whether to use a factored second moment estimator.
+
+  This function returns a tuple with the two largest axes to reduce over.
+  If no two dimensions have size >= min_dim_size_to_factor, return None.
+
+  Args:
+    shape: an input shape
+    factored: whether to use factored second-moment estimator for > 2d vars.
+    min_dim_size_to_factor: only factor accumulator if two array dimensions have at least this size.
+
+  Returns:
+    None or a tuple of ints
+  """
+  if not factored or len(shape) < 2:
+    return None
+  sorted_dims = np.argsort(shape)
+  if shape[sorted_dims[-2]] < min_dim_size_to_factor:
+    return None
+  return int(sorted_dims[-2]), int(sorted_dims[-1])
+
+
+class AdafactorBigVision(Optimizer):
+    """
+    PyTorch implementation of BigVision's Adafactor variant with both single and multi tensor implementations.
+
+
+
+    """
+
+    def __init__(
+            self,
+            params,
+            lr: float = 1.0,
+            min_dim_size_to_factor: int = 32,
+            decay_rate: float = 0.8,
+            decay_offset: int = 0,
+            beta2_cap: float = 0.999,
+            momentum: Optional[float] = 0.9,
+            momentum_dtype: Union[str, torch.dtype] = torch.bfloat16,
+            eps: float = 1e-30,
+            weight_decay: float = 0.0,
+            clipping_threshold: Optional[float] = None,
+            unscaled_wd: bool = False,
+            *,
+            foreach: Optional[bool] = False,
+    ):
+        if isinstance(momentum_dtype, str):
+            if momentum_dtype == 'float16':
+                momentum_dtype = torch.float16
+            elif momentum_dtype == 'bfloat16':
+                momentum_dtype = torch.bfloat16
+            else:
+                assert momentum_dtype == 'float32', f'{momentum_dtype} dtype not supported'
+                momentum_dtype = torch.float32
+
+        defaults = dict(
+            lr=lr,
+            min_dim_size_to_factor=min_dim_size_to_factor,
+            decay_rate=decay_rate,
+            decay_offset=decay_offset,
+            beta2_cap=beta2_cap,
+            momentum=momentum,
+            momentum_dtype=momentum_dtype,
+            eps=eps,
+            weight_decay=weight_decay,
+            clipping_threshold=clipping_threshold,
+            unscaled_wd=unscaled_wd,
+            foreach=foreach,
+        )
+        super().__init__(params, defaults)
+
+    def __setstate__(self, state):
+        super().__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('foreach', None)
+            for p in group['params']:
+                p_state = self.state.get(p, {})
+                if len(p_state) != 0 and not torch.is_tensor(p_state['step']):
+                    p_state['step'] = torch.tensor(float(p_state['step']), dtype=_get_scalar_dtype())
+
+    def _get_beta2(self, step: Tensor, decay_rate: float, beta2_cap: float) -> float:
+        """Computes beta2 according to the step schedule"""
+        t = float(step + 1)
+        return min(beta2_cap, 1.0 - t ** (-decay_rate))
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+
+        for group in self.param_groups:
+            params_with_grad = []
+            grads = []
+            exp_avg_sq_rs = []
+            exp_avg_sq_cs = []
+            exp_avg_sqs = []
+            state_steps = []
+            exp_avgs = []  # For momentum
+
+            for p in group['params']:
+                if p.grad is None:
+                    continue
+
+                if p.grad.is_sparse:
+                    raise RuntimeError("Sparse gradients not supported")
+
+                params_with_grad.append(p)
+                grads.append(p.grad)
+
+                state = self.state[p]
+
+                if len(state) == 0:
+                    # NOTE step on CPU, probably need some more though to make capturable
+                    state['step'] = torch.tensor(0.0, dtype=_get_scalar_dtype())
+
+                    shape = p.grad.shape
+                    factored_dims = _factored_dims(
+                        shape,
+                        factored=True,
+                        min_dim_size_to_factor=self.defaults['min_dim_size_to_factor']
+                    )
+
+                    if factored_dims is not None:
+                        d1, d0 = factored_dims
+                        row_shape = list(p.grad.shape)
+                        row_shape[d0] = 1
+                        col_shape = list(p.grad.shape)
+                        col_shape[d1] = 1
+                        state['exp_avg_sq_r'] = p.grad.new_zeros(row_shape)
+                        state['exp_avg_sq_c'] = p.grad.new_zeros(col_shape)
+                    else:
+                        state['exp_avg_sq'] = torch.zeros_like(p.grad, memory_format=torch.preserve_format)
+
+                    if self.defaults['momentum'] is not None:
+                        state['exp_avg'] = torch.zeros_like(p.grad, dtype=torch.bfloat16)
+
+                state_steps.append(state['step'])
+                exp_avg_sq_rs.append(state.get('exp_avg_sq_r', None))
+                exp_avg_sq_cs.append(state.get('exp_avg_sq_c', None))
+                exp_avg_sqs.append(state.get('exp_avg_sq', None))
+                exp_avgs.append(state.get('exp_avg', None))
+
+            if group['foreach']:
+                func = _multi_tensor_adafactor
+            else:
+                func = _single_tensor_adafactor
+
+            func(
+                params=params_with_grad,
+                grads=grads,
+                exp_avg_sq_rs=exp_avg_sq_rs,
+                exp_avg_sq_cs=exp_avg_sq_cs,
+                exp_avg_sqs=exp_avg_sqs,
+                exp_avgs=exp_avgs,
+                state_steps=state_steps,
+                beta2_decay=group['decay_rate'],
+                beta2_cap=group['beta2_cap'],
+                min_dim_size_to_factor=group['min_dim_size_to_factor'],
+                eps=group['eps'],
+                lr=group['lr'],
+                weight_decay=group['weight_decay'],
+                momentum=group['momentum'],
+                momentum_dtype=group['momentum_dtype'],
+                clipping_threshold=group['clipping_threshold'],
+                unscaled_wd=group['unscaled_wd'],
+            )
+
+        return loss
+
+def _single_tensor_adafactor(
+        params: List[Tensor],
+        grads: List[Tensor],
+        exp_avg_sq_rs: List[Optional[Tensor]],
+        exp_avg_sq_cs: List[Optional[Tensor]],
+        exp_avg_sqs: List[Optional[Tensor]],
+        exp_avgs: List[Optional[Tensor]],
+        state_steps: List[Tensor],
+        *,
+        beta2_decay: float,
+        beta2_cap: float,
+        min_dim_size_to_factor: int,
+        eps: float,
+        lr: float,
+        weight_decay: float,
+        momentum: Optional[float],
+        momentum_dtype: Union[str, torch.dtype],
+        clipping_threshold: Optional[float],
+        unscaled_wd: bool,
+):
+    for i, param in enumerate(params):
+        grad = grads[i]
+        exp_avg_sq_r = exp_avg_sq_rs[i]
+        exp_avg_sq_c = exp_avg_sq_cs[i]
+        exp_avg_sq = exp_avg_sqs[i]
+        exp_avg = exp_avgs[i]
+        step_t = state_steps[i]
+
+        # Update step
+        step_t += 1
+        beta2_t = min(beta2_cap, 1.0 - float(step_t) ** (-beta2_decay))
+        one_minus_beta2_t = 1 - beta2_t
+
+        if exp_avg_sq is None:
+            d1, d0 = _factored_dims(grad.shape, True, min_dim_size_to_factor=min_dim_size_to_factor)
+            grad_sqr = torch.square(grad) + eps
+            exp_avg_sq_r.lerp_(grad_sqr.mean(dim=d0, keepdim=True), one_minus_beta2_t)
+            exp_avg_sq_c.lerp_(grad_sqr.mean(dim=d1, keepdim=True), one_minus_beta2_t)
+
+            reduced_d1 = d1 - 1 if d1 > d0 else d1
+            row_col_mean = exp_avg_sq_r.mean(dim=reduced_d1, keepdim=True)
+            row_factor = (exp_avg_sq_r / row_col_mean).rsqrt()
+            col_factor = exp_avg_sq_c.rsqrt()
+
+            update = grad * row_factor * col_factor
+        else:
+            # Handle non-factored
+            exp_avg_sq.mul_(beta2_t).addcmul_(grad, grad, value=one_minus_beta2_t)
+            update = grad * exp_avg_sq.add(eps).rsqrt_()
+
+        # Clip by RMS value
+        if clipping_threshold is not None:
+            denom = (update.norm(2) / ((update.numel() ** 0.5) / clipping_threshold)).clamp_(max=1.0)
+            update.div_(denom)
+
+        # Apply momentum (in different dtype)
+        if momentum is not None and exp_avg is not None:
+            if momentum_dtype != grad.dtype:
+                exp_avg.lerp_(update.to(momentum_dtype), 1 - momentum)  # ema
+                update = exp_avg.to(grad.dtype)
+            else:
+                exp_avg.lerp_(update, 1 - momentum)  # ema
+                update = exp_avg.clone()
+
+        # Scale by learning rate
+        update.mul_(lr)
+
+        # Perform weight decay
+        if weight_decay != 0:
+            if unscaled_wd:
+                # match big vision impl, 'fully decoupled' decay w/o LR scaling
+                param.mul_(1. - weight_decay)
+            else:
+                # match typical pytorch behaviour for decoupled decay, eg adamw where wd is scaled by LR
+                param.mul_(1. - lr * weight_decay)
+
+        # Update parameters
+        param.add_(update, alpha=-1.0)
+
+def _multi_tensor_adafactor(
+    params: List[Tensor],
+    grads: List[Tensor],
+    exp_avg_sq_rs: List[Optional[Tensor]],
+    exp_avg_sq_cs: List[Optional[Tensor]],
+    exp_avg_sqs: List[Optional[Tensor]],
+    exp_avgs: List[Optional[Tensor]],
+    state_steps: List[Tensor],
+    *,
+    beta2_decay: float,
+    beta2_cap: float,
+    min_dim_size_to_factor: int,
+    eps: float,
+    lr: float,
+    weight_decay: float,
+    momentum: Optional[float],
+    momentum_dtype: Union[str, torch.dtype],
+    clipping_threshold: Optional[float],
+    unscaled_wd: bool,
+):
+    assert False, 'multi-tensor fn (foreach=True) not implemented yet'

timm/optim/optim_factory.py

@@ -10,6 +10,7 @@
 import torch.optim as optim
 
 from timm.models import group_parameters
+from . import AdafactorBigVision
 
 from .adabelief import AdaBelief
 from .adafactor import Adafactor
@@ -356,6 +357,8 @@ def create_optimizer_v2(
     elif opt_lower == 'lion':
         opt_args.pop('eps', None)
         optimizer = Lion(parameters, **opt_args)
+    elif opt_lower == 'adafactorbv':
+        optimizer = AdafactorBigVision(parameters, **opt_args)
 
     # second order
     elif opt_lower == 'adahessian':