[Chore] Update type annotation of input_ids in model forward (#33063)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
@@ -71,7 +71,7 @@ class MyModel(nn.Module):
|
||||
```python
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -36,7 +36,7 @@ class MyGemma2Embedding(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -425,7 +425,7 @@ class AfmoeModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -675,7 +675,7 @@ class AfmoeForCausalLM(nn.Module, SupportsPP, SupportsLoRA):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -542,7 +542,7 @@ class ApertusForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -394,7 +394,7 @@ class ArceeForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -406,7 +406,7 @@ class ArcticModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -460,7 +460,7 @@ class ArcticForCausalLM(nn.Module, SupportsPP, SupportsQuant):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -629,7 +629,7 @@ class AriaForConditionalGeneration(nn.Module, SupportsMultiModal):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -609,7 +609,7 @@ class AudioFlamingo3ForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -420,7 +420,7 @@ class AyaVisionForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsP
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -507,7 +507,7 @@ class BagelForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -311,7 +311,7 @@ class BaiChuanModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -428,7 +428,7 @@ class BaiChuanBaseForCausalLM(nn.Module, SupportsLoRA, SupportsPP, SupportsQuant
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -440,7 +440,7 @@ class BailingMoeModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
position_ids: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -611,7 +611,7 @@ class BailingMoeForCausalLM(nn.Module, SupportsPP, SupportsLoRA):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -311,7 +311,7 @@ class BambaModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -493,7 +493,7 @@ class BambaForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -475,7 +475,7 @@ class BertWithRope(nn.Module, SupportsQuant):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -641,7 +641,7 @@ class Blip2ForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -276,7 +276,7 @@ class BloomModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
position_ids: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -358,7 +358,7 @@ class BloomForCausalLM(nn.Module, SupportsPP, SupportsQuant):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -994,7 +994,7 @@ class ChameleonForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -362,7 +362,7 @@ class ChatGLMModel(nn.Module, SupportsQuant):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -491,7 +491,7 @@ class ChatGLMForCausalLM(ChatGLMBaseModel, SupportsLoRA, SupportsPP, SupportsQua
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -446,7 +446,7 @@ class Cohere2VisionForConditionalGeneration(nn.Module, SupportsMultiModal, Suppo
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -312,7 +312,7 @@ class CohereModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -438,7 +438,7 @@ class CohereForCausalLM(nn.Module, SupportsLoRA, SupportsPP, SupportsQuant):
|
||||
@torch.no_grad()
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -361,7 +361,7 @@ class DbrxModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
position_ids: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -462,7 +462,7 @@ class DbrxForCausalLM(nn.Module, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -213,7 +213,7 @@ class DeepSeekMTP(nn.Module, DeepseekV2MixtureOfExperts):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
hidden_states: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
|
||||
@@ -562,7 +562,7 @@ class DeepseekOCRForCausalLM(nn.Module, SupportsMultiModal, SupportsPP, Supports
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -1085,7 +1085,7 @@ class DeepseekV2Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -1255,7 +1255,7 @@ class DeepseekV2ForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -614,7 +614,7 @@ class DeepseekVLV2ForCausalLM(nn.Module, SupportsMultiModal, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -394,7 +394,7 @@ class Dots1Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -538,7 +538,7 @@ class Dots1ForCausalLM(nn.Module, SupportsPP, SupportsLoRA):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -754,7 +754,7 @@ class DotsOCRForCausalLM(nn.Module, SupportsMultiModal, SupportsPP, SupportsLoRA
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -432,7 +432,7 @@ class Eagle2_5_VLForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -440,7 +440,6 @@ class Eagle2_5_VLForConditionalGeneration(
|
||||
) -> IntermediateTensors:
|
||||
"""Forward pass through the model."""
|
||||
if intermediate_tensors is not None:
|
||||
input_ids = None
|
||||
inputs_embeds = None
|
||||
|
||||
forward_kwargs = {
|
||||
|
||||
@@ -466,7 +466,7 @@ class Ernie4_5_MoeModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -728,7 +728,7 @@ class Ernie4_5_MoeForCausalLM(nn.Module, SupportsPP, SupportsLoRA, MixtureOfExpe
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -1650,7 +1650,7 @@ class Ernie4_5_VLMoeForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -565,7 +565,7 @@ class Ernie4_5_VLMoeModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -646,7 +646,7 @@ class Ernie4_5_VLMoeForCausalLM(nn.Module, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -164,7 +164,7 @@ class ErnieMTP(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
hidden_states: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
|
||||
@@ -496,7 +496,7 @@ class ExaoneForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -490,7 +490,7 @@ class Exaone4ForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -549,7 +549,7 @@ class ExaoneMoeForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -402,7 +402,7 @@ class FalconModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -459,7 +459,7 @@ class FalconH1Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -602,7 +602,7 @@ class FalconH1ForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -340,7 +340,7 @@ class FuyuForCausalLM(nn.Module, SupportsMultiModal, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -297,7 +297,7 @@ class GemmaModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -400,7 +400,7 @@ class GemmaForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -410,7 +410,7 @@ class Gemma2ForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -494,7 +494,7 @@ class Gemma3ForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -618,7 +618,7 @@ class Gemma3ForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -704,7 +704,7 @@ class Gemma3nSelfDecoder(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
per_layer_inputs: torch.Tensor | None = None,
|
||||
@@ -887,7 +887,7 @@ class Gemma3nTextModel(nn.Module, SupportsQuant):
|
||||
|
||||
def fast_prefill_forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
per_layer_inputs: torch.Tensor | None = None,
|
||||
@@ -964,7 +964,7 @@ class Gemma3nTextModel(nn.Module, SupportsQuant):
|
||||
|
||||
def normal_forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
per_layer_inputs: torch.Tensor | None = None,
|
||||
@@ -1131,7 +1131,7 @@ class Gemma3nForCausalLM(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
*,
|
||||
per_layer_inputs: torch.Tensor | None = None,
|
||||
|
||||
@@ -707,7 +707,7 @@ class Gemma3nForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -270,7 +270,7 @@ class Glm4ForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -1711,7 +1711,7 @@ class Glm4vForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -451,7 +451,7 @@ class Glm4MoeModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -687,7 +687,7 @@ class Glm4MoeForCausalLM(nn.Module, SupportsPP, SupportsLoRA, Glm4MixtureOfExper
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -264,7 +264,7 @@ class Glm4MoeLiteModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -596,7 +596,7 @@ class Glm4MoeLiteForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -230,7 +230,7 @@ class Glm4MoeLiteMTP(nn.Module, SupportsPP, Glm4MixtureOfExperts):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
hidden_states: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
|
||||
@@ -216,7 +216,7 @@ class Glm4MoeMTP(nn.Module, Glm4MixtureOfExperts):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
hidden_states: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
|
||||
@@ -769,7 +769,7 @@ class GLM4VForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -1075,7 +1075,7 @@ class GlmAsrForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -218,7 +218,7 @@ class GPT2Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
position_ids: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None,
|
||||
@@ -298,7 +298,7 @@ class GPT2LMHeadModel(nn.Module, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -362,7 +362,7 @@ class GPT2ForSequenceClassification(nn.Module, SupportsCrossEncoding):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -235,7 +235,7 @@ class GPTBigCodeModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
position_ids: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -311,7 +311,7 @@ class GPTBigCodeForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -220,7 +220,7 @@ class GPTJModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
position_ids: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -324,7 +324,7 @@ class GPTJForCausalLM(nn.Module, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -230,7 +230,7 @@ class GPTNeoXModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
position_ids: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -318,7 +318,7 @@ class GPTNeoXForCausalLM(nn.Module, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -275,7 +275,7 @@ class GptOssModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -714,7 +714,7 @@ class GptOssForCausalLM(nn.Module, SupportsPP, SupportsEagle3, SupportsLoRA):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -437,7 +437,7 @@ class GraniteForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -806,7 +806,7 @@ class GraniteSpeechForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -312,7 +312,7 @@ class GraniteMoeModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -528,7 +528,7 @@ class GraniteMoeForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -368,7 +368,7 @@ class GraniteMoeHybridModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -685,7 +685,7 @@ class GraniteMoeHybridForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -183,7 +183,7 @@ class GraniteMoeSharedModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -295,7 +295,7 @@ class GraniteMoeSharedForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -491,7 +491,7 @@ class Grok1Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -705,7 +705,7 @@ class GrokBaseForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -938,7 +938,7 @@ class HunyuanV1ModelBase(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -968,7 +968,7 @@ class HunYuanVLForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None,
|
||||
|
||||
@@ -747,7 +747,7 @@ class HCXVisionForCausalLM(nn.Module, SupportsMultiModal, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -559,7 +559,7 @@ class Idefics3Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -689,7 +689,7 @@ class Idefics3ForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsLo
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -603,6 +603,8 @@ class SupportsPP(Protocol):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: Tensor | None,
|
||||
positions: Tensor,
|
||||
*,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
) -> IntermediateTensors | None:
|
||||
@@ -631,6 +633,8 @@ class _SupportsPPType(Protocol):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: Tensor | None,
|
||||
positions: Tensor,
|
||||
*,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
) -> Tensor | IntermediateTensors: ...
|
||||
|
||||
@@ -284,7 +284,7 @@ class InternLM2Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -350,7 +350,7 @@ class InternLM2ForCausalLM(nn.Module, SupportsPP, SupportsLoRA):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -446,7 +446,7 @@ class InternLM2ForRewardModel(InternLM2ForCausalLM):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -101,7 +101,7 @@ class InternLM2VEModel(InternLM2Model):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -782,7 +782,7 @@ class InternS1ForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -1371,7 +1371,7 @@ class InternVLChatModel(nn.Module, SupportsMultiModal, SupportsPP, SupportsLoRA)
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -438,7 +438,7 @@ class IQuestLoopCoderModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -570,7 +570,7 @@ class IQuestLoopCoderForCausalLM(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -1450,7 +1450,7 @@ class IsaacForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -280,7 +280,7 @@ class JAISModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
position_ids: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -344,7 +344,7 @@ class JAISLMHeadModel(nn.Module, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -483,7 +483,7 @@ class Jais2ForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -348,7 +348,7 @@ class JambaModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -516,7 +516,7 @@ class JambaForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -125,7 +125,7 @@ class JinaVLForSequenceClassification(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -732,7 +732,7 @@ class KananaVForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsPP)
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -1438,7 +1438,7 @@ class BaseKeyeModule(nn.Module, SupportsMultiModal):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -506,7 +506,7 @@ class KimiLinearForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -389,7 +389,7 @@ class KimiVLForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -342,7 +342,7 @@ class Lfm2Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -503,7 +503,7 @@ class Lfm2ForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -457,7 +457,7 @@ class Lfm2MoeModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -730,7 +730,7 @@ class Lfm2MoeForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -769,7 +769,7 @@ class Lfm2VLForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -581,7 +581,7 @@ class LlamaForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -662,7 +662,7 @@ class LlavaForConditionalGeneration(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -509,7 +509,7 @@ class LlavaNextForConditionalGeneration(nn.Module, SupportsMultiModal, SupportsP
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -426,7 +426,7 @@ class LlavaNextVideoForConditionalGeneration(nn.Module, SupportsMultiModal, Supp
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -887,7 +887,7 @@ class LlavaOnevisionForConditionalGeneration(nn.Module, SupportsMultiModal, Supp
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -520,7 +520,7 @@ class FlashModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -605,7 +605,7 @@ class LongcatFlashForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -150,7 +150,7 @@ class LongCatFlashMTP(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
hidden_states: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
|
||||
@@ -142,7 +142,7 @@ class MambaModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -225,7 +225,7 @@ class MambaForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -137,7 +137,7 @@ class Mamba2Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -267,7 +267,7 @@ class Mamba2ForCausalLM(
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -796,7 +796,7 @@ class MiDashengLMModel(nn.Module, SupportsMultiModal, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -61,7 +61,7 @@ logger = init_logger(__name__)
|
||||
class MiMoModel(Qwen2Model):
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -169,7 +169,7 @@ class MiMoMTP(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
hidden_states: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
|
||||
@@ -478,7 +478,7 @@ class MiMoV2Model(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -693,7 +693,7 @@ class MiMoV2FlashForCausalLM(nn.Module, SupportsPP, MixtureOfExperts):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -440,7 +440,7 @@ class MiniCPMModel(nn.Module):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
@@ -620,7 +620,7 @@ class MiniCPMForCausalLM(nn.Module, SupportsLoRA, SupportsPP, SupportsEagle3):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
@@ -1147,7 +1147,7 @@ class MiniCPMVBaseModel(nn.Module, SupportsMultiModal, SupportsPP):
|
||||
|
||||
def forward(
|
||||
self,
|
||||
input_ids: torch.Tensor,
|
||||
input_ids: torch.Tensor | None,
|
||||
positions: torch.Tensor,
|
||||
intermediate_tensors: IntermediateTensors | None = None,
|
||||
inputs_embeds: torch.Tensor | None = None,
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user