Update deprecated type hinting in models (#18132)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2025-05-14 22:06:50 -07:00
committed by GitHub
parent 83f74c698f
commit 26d0419309
130 changed files with 971 additions and 901 deletions
+7 -6
View File
@@ -23,7 +23,8 @@
# limitations under the License.
"""Inference-only Solar model compatible with HuggingFace weights."""
from typing import Any, Dict, Iterable, Optional, Set, Tuple, Union
from collections.abc import Iterable
from typing import Any, Optional, Union
import torch
from torch import nn
@@ -101,7 +102,7 @@ class SolarAttention(nn.Module):
num_heads: int,
num_kv_heads: int,
rope_theta: float = 10000,
rope_scaling: Optional[Dict[str, Any]] = None,
rope_scaling: Optional[dict[str, Any]] = None,
max_position_embeddings: int = 8192,
quant_config: Optional[QuantizationConfig] = None,
bias: bool = False,
@@ -236,7 +237,7 @@ class SolarDecoderLayer(nn.Module):
positions: torch.Tensor,
hidden_states: torch.Tensor,
residual: Optional[torch.Tensor],
) -> Tuple[torch.Tensor, torch.Tensor]:
) -> tuple[torch.Tensor, torch.Tensor]:
# Self Attention
if residual is None:
residual = hidden_states
@@ -437,8 +438,8 @@ class SolarForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
sampling_metadata)
return logits
def load_weights(self, weights: Iterable[Tuple[str,
torch.Tensor]]) -> Set[str]:
def load_weights(self, weights: Iterable[tuple[str,
torch.Tensor]]) -> set[str]:
stacked_params_mapping = [
# (param_name, shard_name, shard_id)
(".qkv_proj", ".q_proj", "q"),
@@ -448,7 +449,7 @@ class SolarForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
(".gate_up_proj", ".up_proj", 1),
]
params_dict = dict(self.named_parameters())
loaded_params: Set[str] = set()
loaded_params: set[str] = set()
for name, loaded_weight in weights:
if "rotary_emb.inv_freq" in name:
continue