daking's picture
LLM-foundry update October 17, 2023 23:04:30 (#4)
484f35e
raw
history blame
3.12 kB
from typing import Dict, List, Optional, Type, Union
import torch
def _cast_if_autocast_enabled(tensor: torch.Tensor) -> torch.Tensor:
if torch.is_autocast_enabled():
if tensor.device.type == 'cuda':
dtype = torch.get_autocast_gpu_dtype()
elif tensor.device.type == 'cpu':
dtype = torch.get_autocast_cpu_dtype()
else:
raise NotImplementedError()
return tensor.to(dtype=dtype)
return tensor
class LPLayerNorm(torch.nn.LayerNorm):
def __init__(self, normalized_shape: Union[int, List[int], torch.Size], eps: float=1e-05, elementwise_affine: bool=True, device: Optional[torch.device]=None, dtype: Optional[torch.dtype]=None):
super().__init__(normalized_shape=normalized_shape, eps=eps, elementwise_affine=elementwise_affine, device=device, dtype=dtype)
def forward(self, x: torch.Tensor) -> torch.Tensor:
module_device = x.device
downcast_x = _cast_if_autocast_enabled(x)
downcast_weight = _cast_if_autocast_enabled(self.weight) if self.weight is not None else self.weight
downcast_bias = _cast_if_autocast_enabled(self.bias) if self.bias is not None else self.bias
with torch.autocast(enabled=False, device_type=module_device.type):
return torch.nn.functional.layer_norm(downcast_x, self.normalized_shape, downcast_weight, downcast_bias, self.eps)
def rms_norm(x: torch.Tensor, weight: Optional[torch.Tensor]=None, eps: float=1e-05) -> torch.Tensor:
output = x * torch.rsqrt(x.pow(2).mean(-1, keepdim=True) + eps)
if weight is not None:
return output * weight
return output
class RMSNorm(torch.nn.Module):
def __init__(self, normalized_shape: Union[int, List[int], torch.Size], eps: float=1e-05, weight: bool=True, dtype: Optional[torch.dtype]=None, device: Optional[torch.device]=None):
super().__init__()
self.eps = eps
if weight:
self.weight = torch.nn.Parameter(torch.ones(normalized_shape, dtype=dtype, device=device))
else:
self.register_parameter('weight', None)
def forward(self, x: torch.Tensor) -> torch.Tensor:
return rms_norm(x.float(), self.weight, self.eps).to(dtype=x.dtype)
class LPRMSNorm(RMSNorm):
def __init__(self, normalized_shape: Union[int, List[int], torch.Size], eps: float=1e-05, weight: bool=True, dtype: Optional[torch.dtype]=None, device: Optional[torch.device]=None):
super().__init__(normalized_shape=normalized_shape, eps=eps, weight=weight, dtype=dtype, device=device)
def forward(self, x: torch.Tensor) -> torch.Tensor:
downcast_x = _cast_if_autocast_enabled(x)
downcast_weight = _cast_if_autocast_enabled(self.weight) if self.weight is not None else self.weight
with torch.autocast(enabled=False, device_type=x.device.type):
return rms_norm(downcast_x, downcast_weight, self.eps).to(dtype=x.dtype)
NORM_CLASS_REGISTRY: Dict[str, Type[torch.nn.Module]] = {'layernorm': torch.nn.LayerNorm, 'low_precision_layernorm': LPLayerNorm, 'rmsnorm': RMSNorm, 'low_precision_rmsnorm': LPRMSNorm}