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:
@@ -10,7 +10,7 @@
|
||||
|
||||
from collections.abc import Iterable
|
||||
from itertools import islice
|
||||
from typing import Any, Optional, Union
|
||||
from typing import Any
|
||||
|
||||
import torch
|
||||
from torch import nn
|
||||
@@ -52,7 +52,7 @@ class ArceeMLP(nn.Module):
|
||||
hidden_size: int,
|
||||
intermediate_size: int,
|
||||
hidden_act: str,
|
||||
quant_config: Optional[Any] = None,
|
||||
quant_config: Any | None = None,
|
||||
bias: bool = False,
|
||||
prefix: str = "",
|
||||
reduce_results: bool = True,
|
||||
@@ -98,8 +98,8 @@ class ArceeDecoderLayer(nn.Module):
|
||||
def __init__(
|
||||
self,
|
||||
config: LlamaConfig,
|
||||
cache_config: Optional[Any] = None,
|
||||
quant_config: Optional[Any] = None,
|
||||
cache_config: Any | None = None,
|
||||
quant_config: Any | None = None,
|
||||
prefix: str = "",
|
||||
) -> None:
|
||||
super().__init__()
|
||||
@@ -165,7 +165,7 @@ class ArceeDecoderLayer(nn.Module):
|
||||
self,
|
||||
positions: torch.Tensor,
|
||||
hidden_states: torch.Tensor,
|
||||
residual: Optional[torch.Tensor],
|
||||
residual: torch.Tensor | None,
|
||||
) -> tuple[torch.Tensor, torch.Tensor]:
|
||||
# Self-Attention block
|
||||
if residual is None:
|
||||
@@ -247,13 +247,11 @@ class ArceeModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: Optional[torch.Tensor],
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: Optional[IntermediateTensors],
|
||||
inputs_embeds: Optional[torch.Tensor] = None,
|
||||
) -> Union[
|
||||
torch.Tensor, IntermediateTensors, tuple[torch.Tensor, list[torch.Tensor]]
|
||||
]:
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
) -> torch.Tensor | IntermediateTensors | tuple[torch.Tensor, list[torch.Tensor]]:
|
||||
# Embedding lookup (on first pipeline rank)
|
||||
if get_pp_group().is_first_rank:
|
||||
hidden_states = (
|
||||
@@ -415,9 +413,9 @@ class ArceeForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
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:
|
||||
model_output = self.model(
|
||||
input_ids=input_ids,
|
||||
positions=positions,
|
||||
@@ -426,7 +424,7 @@ class ArceeForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
)
|
||||
return model_output
|
||||
|
||||
def compute_logits(self, hidden_states: torch.Tensor) -> Optional[torch.Tensor]:
|
||||
def compute_logits(self, hidden_states: torch.Tensor) -> torch.Tensor | None:
|
||||
# Compute final logits from hidden states (last pipeline rank only)
|
||||
logits = self.logits_processor(self.lm_head, hidden_states)
|
||||
return logits
|
||||
|
||||
Reference in New Issue
Block a user