Update Optional[x] -> x | None and Union[x, y] to x | y (#26633)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
@@ -27,7 +27,6 @@
|
||||
import typing
|
||||
from collections.abc import Callable, Iterable
|
||||
from itertools import islice
|
||||
from typing import Optional, Union
|
||||
|
||||
import torch
|
||||
from torch import nn
|
||||
@@ -88,10 +87,10 @@ class MixtralMoE(nn.Module):
|
||||
top_k: int,
|
||||
hidden_size: int,
|
||||
intermediate_size: int,
|
||||
params_dtype: Optional[torch.dtype] = None,
|
||||
quant_config: Optional[QuantizationConfig] = None,
|
||||
tp_size: Optional[int] = None,
|
||||
dp_size: Optional[int] = None,
|
||||
params_dtype: torch.dtype | None = None,
|
||||
quant_config: QuantizationConfig | None = None,
|
||||
tp_size: int | None = None,
|
||||
dp_size: int | None = None,
|
||||
prefix: str = "",
|
||||
enable_eplb: bool = False,
|
||||
):
|
||||
@@ -163,8 +162,8 @@ class MixtralAttention(nn.Module):
|
||||
num_kv_heads: int,
|
||||
max_position: int = 4096 * 32,
|
||||
rope_theta: float = 10000,
|
||||
cache_config: Optional[CacheConfig] = None,
|
||||
quant_config: Optional[QuantizationConfig] = None,
|
||||
cache_config: CacheConfig | None = None,
|
||||
quant_config: QuantizationConfig | None = None,
|
||||
prefix: str = "",
|
||||
) -> None:
|
||||
super().__init__()
|
||||
@@ -242,8 +241,8 @@ class MixtralDecoderLayer(nn.Module):
|
||||
def __init__(
|
||||
self,
|
||||
config: MixtralConfig,
|
||||
cache_config: Optional[CacheConfig] = None,
|
||||
quant_config: Optional[QuantizationConfig] = None,
|
||||
cache_config: CacheConfig | None = None,
|
||||
quant_config: QuantizationConfig | None = None,
|
||||
prefix: str = "",
|
||||
enable_eplb: bool = False,
|
||||
) -> None:
|
||||
@@ -280,7 +279,7 @@ class MixtralDecoderLayer(nn.Module):
|
||||
self,
|
||||
positions: torch.Tensor,
|
||||
hidden_states: torch.Tensor,
|
||||
residual: Optional[torch.Tensor],
|
||||
residual: torch.Tensor | None,
|
||||
) -> torch.Tensor:
|
||||
# Self Attention
|
||||
if residual is None:
|
||||
@@ -353,9 +352,9 @@ class MixtralModel(nn.Module):
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: Optional[IntermediateTensors],
|
||||
inputs_embeds: Optional[torch.Tensor] = None,
|
||||
) -> Union[torch.Tensor, IntermediateTensors]:
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
) -> torch.Tensor | IntermediateTensors:
|
||||
if get_pp_group().is_first_rank:
|
||||
if inputs_embeds is not None:
|
||||
hidden_states = inputs_embeds
|
||||
@@ -615,9 +614,9 @@ class MixtralForCausalLM(nn.Module, SupportsLoRA, SupportsPP, MixtureOfExperts):
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: Optional[IntermediateTensors] = None,
|
||||
inputs_embeds: Optional[torch.Tensor] = None,
|
||||
) -> Union[torch.Tensor, IntermediateTensors]:
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
) -> torch.Tensor | IntermediateTensors:
|
||||
hidden_states = self.model(
|
||||
input_ids, positions, intermediate_tensors, inputs_embeds
|
||||
)
|
||||
@@ -626,7 +625,7 @@ class MixtralForCausalLM(nn.Module, SupportsLoRA, SupportsPP, MixtureOfExperts):
|
||||
def compute_logits(
|
||||
self,
|
||||
hidden_states: torch.Tensor,
|
||||
) -> Optional[torch.Tensor]:
|
||||
) -> torch.Tensor | None:
|
||||
logits = self.logits_processor(self.lm_head, hidden_states)
|
||||
return logits
|
||||
|
||||
|
||||
Reference in New Issue
Block a user