biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Models] Add remaining model PP support (#7168)

Browse Source 
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:
Murali Andoorveedu
2024-10-03 19:56:58 -07:00
committed by GitHub
parent 303d44790a
commit 0f6d7a9a34
69 changed files with 2585 additions and 1344 deletions
Download Patch File Download Diff File Expand all files Collapse all files

60
vllm/model_executor/models/orion.py
View File

@@ -4,7 +4,7 @@
# Copyright (c) OrionStar Inc.
# LICENSE: https://huggingface.co/OrionStarAI/Orion-14B-Base/blob/main/LICENSE
"""Inference-only Orion-14B model compatible with HuggingFace weights."""
from typing import Any, Dict, Iterable, List, Optional, Tuple
from typing import Any, Dict, Iterable, List, Optional, Tuple, Union
import torch
from torch import nn
@@ -12,14 +12,13 @@ from transformers import PretrainedConfig
from vllm.attention import Attention, AttentionMetadata
from vllm.config import CacheConfig
from vllm.distributed import get_tensor_model_parallel_world_size
from vllm.distributed import get_pp_group, get_tensor_model_parallel_world_size
from vllm.model_executor.layers.activation import SiluAndMul
from vllm.model_executor.layers.linear import (MergedColumnParallelLinear,
QKVParallelLinear,
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 (
@@ -28,6 +27,10 @@ from vllm.model_executor.model_loader.weight_utils import default_weight_loader
from vllm.model_executor.sampling_metadata import SamplingMetadata
from vllm.sequence import IntermediateTensors
from .interfaces import SupportsPP
from .utils import (is_pp_missing_parameter,
make_empty_intermediate_tensors_factory, make_layers)
class OrionMLP(nn.Module):
@@ -210,6 +213,7 @@ class OrionModel(nn.Module):
config: PretrainedConfig,
cache_config: Optional[CacheConfig] = None,
quant_config: Optional[QuantizationConfig] = None,
prefix: str = "",
) -> None:
super().__init__()
self.config = config
@@ -219,11 +223,18 @@ class OrionModel(nn.Module):
config.vocab_size,
config.hidden_size,
)
self.layers = nn.ModuleList([
OrionDecoderLayer(config, cache_config, quant_config)
for _ in range(config.num_hidden_layers)
])
self.start_layer, self.end_layer, self.layers = make_layers(
config.num_hidden_layers,
lambda prefix: OrionDecoderLayer(
config,
cache_config,
quant_config,
),
prefix=f"{prefix}.layers")
self.norm = nn.LayerNorm(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,
@@ -231,23 +242,34 @@ class OrionModel(nn.Module):
positions: torch.Tensor,
kv_caches: List[torch.Tensor],
attn_metadata: AttentionMetadata,
) -> torch.Tensor:
hidden_states = self.embed_tokens(input_ids)
residual = None
for i in range(len(self.layers)):
intermediate_tensors: Optional[IntermediateTensors],
) -> Union[torch.Tensor, IntermediateTensors]:
if get_pp_group().is_first_rank:
hidden_states = self.embed_tokens(input_ids)
residual = None
else:
assert intermediate_tensors
hidden_states = intermediate_tensors["hidden_states"]
residual = intermediate_tensors["residual"]
for i in range(self.start_layer, self.end_layer):
layer = self.layers[i]
hidden_states, residual = layer(
positions,
hidden_states,
kv_caches[i],
kv_caches[i - self.start_layer],
attn_metadata,
residual,
)
if not get_pp_group().is_last_rank:
return IntermediateTensors({
"hidden_states": hidden_states,
"residual": residual
})
hidden_states = self.norm(hidden_states)
return hidden_states
class OrionForCausalLM(nn.Module):
class OrionForCausalLM(nn.Module, SupportsPP):
def __init__(
self,
@@ -266,6 +288,8 @@ class OrionForCausalLM(nn.Module):
self.lm_head.weight = self.model.embed_tokens.weight
self.logits_processor = LogitsProcessor(config.vocab_size)
self.sampler = Sampler()
self.make_empty_intermediate_tensors = (
self.model.make_empty_intermediate_tensors)
def forward(
self,
@@ -274,9 +298,9 @@ class OrionForCausalLM(nn.Module):
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)
attn_metadata, intermediate_tensors)
return hidden_states
def compute_logits(
@@ -321,6 +345,8 @@ class OrionForCausalLM(nn.Module):
# Skip loading extra bias for GPTQ models.
if name.endswith(".bias") and name not in params_dict:
continue
if is_pp_missing_parameter(name, self):
continue
param = params_dict[name]
weight_loader = param.weight_loader
weight_loader(param, loaded_weight, shard_id)
@@ -329,6 +355,8 @@ class OrionForCausalLM(nn.Module):
# Skip loading extra bias for GPTQ models.
if name.endswith(".bias") and name not in params_dict:
continue
if is_pp_missing_parameter(name, self):
continue
param = params_dict[name]
weight_loader = getattr(param, "weight_loader",
default_weight_loader)