[Models] Add remaining model PP support (#7168)
Signed-off-by: Muralidhar Andoorveedu <muralidhar.andoorveedu@centml.ai> Signed-off-by: Murali Andoorveedu <muralidhar.andoorveedu@centml.ai> Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
co-authored by
DarkLight1337
parent
303d44790a
commit
0f6d7a9a34
@@ -21,7 +21,7 @@
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
"""Inference-only Mixtral model."""
|
||||
from typing import Iterable, List, Optional, Tuple
|
||||
from typing import Iterable, List, Optional, Tuple, Union
|
||||
|
||||
import torch
|
||||
from torch import nn
|
||||
@@ -36,8 +36,7 @@ from vllm.model_executor.layers.linear import (QKVParallelLinear,
|
||||
ReplicatedLinear,
|
||||
RowParallelLinear)
|
||||
from vllm.model_executor.layers.logits_processor import LogitsProcessor
|
||||
from vllm.model_executor.layers.quantization.base_config import (
|
||||
QuantizationConfig)
|
||||
from vllm.model_executor.layers.quantization import QuantizationConfig
|
||||
from vllm.model_executor.layers.rotary_embedding import get_rope
|
||||
from vllm.model_executor.layers.sampler import Sampler, SamplerOutput
|
||||
from vllm.model_executor.layers.vocab_parallel_embedding import (
|
||||
@@ -47,8 +46,9 @@ from vllm.model_executor.model_loader.weight_utils import (
|
||||
from vllm.model_executor.sampling_metadata import SamplingMetadata
|
||||
from vllm.sequence import IntermediateTensors
|
||||
|
||||
from .interfaces import SupportsLoRA
|
||||
from .utils import is_pp_missing_parameter, make_layers
|
||||
from .interfaces import SupportsLoRA, SupportsPP
|
||||
from .utils import (is_pp_missing_parameter,
|
||||
make_empty_intermediate_tensors_factory, make_layers)
|
||||
|
||||
|
||||
class MixtralMoE(nn.Module):
|
||||
@@ -276,6 +276,9 @@ class MixtralModel(nn.Module):
|
||||
prefix=f"{prefix}.layers")
|
||||
|
||||
self.norm = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
|
||||
self.make_empty_intermediate_tensors = (
|
||||
make_empty_intermediate_tensors_factory(
|
||||
["hidden_states", "residual"], config.hidden_size))
|
||||
|
||||
def forward(
|
||||
self,
|
||||
@@ -284,7 +287,7 @@ class MixtralModel(nn.Module):
|
||||
kv_caches: List[torch.Tensor],
|
||||
attn_metadata: AttentionMetadata,
|
||||
intermediate_tensors: Optional[IntermediateTensors],
|
||||
) -> torch.Tensor:
|
||||
) -> Union[torch.Tensor, IntermediateTensors]:
|
||||
if get_pp_group().is_first_rank:
|
||||
hidden_states = self.embed_tokens(input_ids)
|
||||
residual = None
|
||||
@@ -306,7 +309,7 @@ class MixtralModel(nn.Module):
|
||||
return hidden_states
|
||||
|
||||
|
||||
class MixtralForCausalLM(nn.Module, SupportsLoRA):
|
||||
class MixtralForCausalLM(nn.Module, SupportsLoRA, SupportsPP):
|
||||
fall_back_to_pt_during_load = False
|
||||
|
||||
packed_modules_mapping = {
|
||||
@@ -365,6 +368,8 @@ class MixtralForCausalLM(nn.Module, SupportsLoRA):
|
||||
self.logits_processor = LogitsProcessor(self.unpadded_vocab_size,
|
||||
config.vocab_size)
|
||||
self.sampler = Sampler()
|
||||
self.make_empty_intermediate_tensors = (
|
||||
self.model.make_empty_intermediate_tensors)
|
||||
|
||||
def forward(
|
||||
self,
|
||||
@@ -373,7 +378,7 @@ class MixtralForCausalLM(nn.Module, SupportsLoRA):
|
||||
kv_caches: List[torch.Tensor],
|
||||
attn_metadata: AttentionMetadata,
|
||||
intermediate_tensors: Optional[IntermediateTensors] = None,
|
||||
) -> torch.Tensor:
|
||||
) -> Union[torch.Tensor, IntermediateTensors]:
|
||||
hidden_states = self.model(input_ids, positions, kv_caches,
|
||||
attn_metadata, intermediate_tensors)
|
||||
return hidden_states
|
||||
@@ -387,20 +392,6 @@ class MixtralForCausalLM(nn.Module, SupportsLoRA):
|
||||
sampling_metadata)
|
||||
return logits
|
||||
|
||||
def make_empty_intermediate_tensors(
|
||||
self, batch_size: int, dtype: torch.dtype,
|
||||
device: torch.device) -> IntermediateTensors:
|
||||
return IntermediateTensors({
|
||||
"hidden_states":
|
||||
torch.zeros((batch_size, self.config.hidden_size),
|
||||
dtype=dtype,
|
||||
device=device),
|
||||
"residual":
|
||||
torch.zeros((batch_size, self.config.hidden_size),
|
||||
dtype=dtype,
|
||||
device=device),
|
||||
})
|
||||
|
||||
def sample(
|
||||
self,
|
||||
logits: Optional[torch.Tensor],
|
||||
|
||||
Reference in New Issue
Block a user