From 63a164172dbcc43857dbcf6443a7594faa143151 Mon Sep 17 00:00:00 2001 From: youkaichao Date: Sun, 1 Dec 2024 19:27:13 -0800 Subject: [PATCH] [misc] remove xverse modeling file (#10814) Signed-off-by: youkaichao --- vllm/model_executor/models/registry.py | 2 +- vllm/model_executor/models/xverse.py | 423 ------------------------- 2 files changed, 1 insertion(+), 424 deletions(-) delete mode 100644 vllm/model_executor/models/xverse.py diff --git a/vllm/model_executor/models/registry.py b/vllm/model_executor/models/registry.py index 2b7b69e8c3a95..c66fbce018a62 100644 --- a/vllm/model_executor/models/registry.py +++ b/vllm/model_executor/models/registry.py @@ -94,7 +94,7 @@ "Starcoder2ForCausalLM": ("starcoder2", "Starcoder2ForCausalLM"), "SolarForCausalLM": ("solar", "SolarForCausalLM"), "TeleChat2ForCausalLM": ("telechat2", "TeleChat2ForCausalLM"), - "XverseForCausalLM": ("xverse", "XverseForCausalLM"), + "XverseForCausalLM": ("llama", "LlamaForCausalLM"), # [Encoder-decoder] "BartModel": ("bart", "BartForConditionalGeneration"), "BartForConditionalGeneration": ("bart", "BartForConditionalGeneration"), diff --git a/vllm/model_executor/models/xverse.py b/vllm/model_executor/models/xverse.py deleted file mode 100644 index 25a0d474e2863..0000000000000 --- a/vllm/model_executor/models/xverse.py +++ /dev/null @@ -1,423 +0,0 @@ -# Adapted from -# https://huggingface.co/xverse/XVERSE-7B/blob/main/modeling_xverse.py -# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved. -# -# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX -# and OPT implementations in this library. It has been modified from its -# original forms to accommodate minor architectural differences compared -# to GPT-NeoX and OPT used by the Meta AI team that trained the model. -# -# Licensed under the Apache License, Version 2.0 (the "License"); -# you may not use this file except in compliance with the License. -# You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -"""Inference-only Xverse model compatible with HuggingFace weights.""" -from typing import Any, Dict, Iterable, List, Optional, Set, Tuple, Union - -import torch -from torch import nn -from transformers import PretrainedConfig - -from vllm.attention import Attention, AttentionMetadata -from vllm.compilation.decorators import support_torch_compile -from vllm.config import CacheConfig, VllmConfig -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.layernorm import RMSNorm -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 import QuantizationConfig -from vllm.model_executor.layers.rotary_embedding import get_rope -from vllm.model_executor.layers.sampler import SamplerOutput, get_sampler -from vllm.model_executor.layers.vocab_parallel_embedding import ( - ParallelLMHead, VocabParallelEmbedding) -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 SupportsLoRA, SupportsPP -from .utils import (is_pp_missing_parameter, - make_empty_intermediate_tensors_factory, make_layers, - maybe_prefix) - - -class XverseMLP(nn.Module): - - def __init__( - self, - hidden_size: int, - intermediate_size: int, - hidden_act: str, - quant_config: Optional[QuantizationConfig] = None, - ) -> None: - super().__init__() - self.gate_up_proj = MergedColumnParallelLinear( - hidden_size, [intermediate_size] * 2, - bias=False, - quant_config=quant_config) - self.down_proj = RowParallelLinear(intermediate_size, - hidden_size, - bias=False, - quant_config=quant_config) - if hidden_act != "silu": - raise ValueError(f"Unsupported activation: {hidden_act}. " - "Only silu is supported for now.") - self.act_fn = SiluAndMul() - - def forward(self, x): - gate, _ = self.gate_up_proj(x) - x = self.act_fn(gate) - x, _ = self.down_proj(x) - return x - - -class XverseAttention(nn.Module): - - def __init__( - self, - hidden_size: int, - num_heads: int, - num_kv_heads: int, - rope_theta: float = 10000, - rope_scaling: Optional[Dict[str, Any]] = None, - max_position_embeddings: int = 8192, - quant_config: Optional[QuantizationConfig] = None, - bias: bool = False, - cache_config: Optional[CacheConfig] = None, - prefix: str = "", - ) -> None: - super().__init__() - self.hidden_size = hidden_size - tp_size = get_tensor_model_parallel_world_size() - self.total_num_heads = num_heads - assert self.total_num_heads % tp_size == 0 - self.num_heads = self.total_num_heads // tp_size - self.total_num_kv_heads = num_kv_heads - # partition the KV heads across multiple tensor parallel GPUs. - assert self.total_num_kv_heads % tp_size == 0 - self.num_kv_heads = max(1, self.total_num_kv_heads // tp_size) - self.head_dim = hidden_size // self.total_num_heads - self.q_size = self.num_heads * self.head_dim - self.kv_size = self.num_kv_heads * self.head_dim - self.scaling = self.head_dim**-0.5 - self.rope_theta = rope_theta - self.max_position_embeddings = max_position_embeddings - - self.qkv_proj = QKVParallelLinear( - hidden_size, - self.head_dim, - self.total_num_heads, - self.total_num_kv_heads, - bias=bias, - quant_config=quant_config, - ) - self.o_proj = RowParallelLinear( - self.total_num_heads * self.head_dim, - hidden_size, - bias=bias, - quant_config=quant_config, - ) - - self.rotary_emb = get_rope( - self.head_dim, - rotary_dim=self.head_dim, - max_position=max_position_embeddings, - base=rope_theta, - rope_scaling=rope_scaling, - ) - self.attn = Attention(self.num_heads, - self.head_dim, - self.scaling, - num_kv_heads=self.num_kv_heads, - cache_config=cache_config, - quant_config=quant_config, - prefix=f"{prefix}.attn") - - def forward( - self, - positions: torch.Tensor, - hidden_states: torch.Tensor, - kv_cache: torch.Tensor, - attn_metadata: AttentionMetadata, - ) -> torch.Tensor: - qkv, _ = self.qkv_proj(hidden_states) - q, k, v = qkv.split([self.q_size, self.kv_size, self.kv_size], dim=-1) - q, k = self.rotary_emb(positions, q, k) - attn_output = self.attn(q, k, v, kv_cache, attn_metadata) - output, _ = self.o_proj(attn_output) - return output - - -class XverseDecoderLayer(nn.Module): - - def __init__( - self, - config: PretrainedConfig, - cache_config: Optional[CacheConfig] = None, - quant_config: Optional[QuantizationConfig] = None, - prefix: str = "", - ) -> None: - super().__init__() - self.hidden_size = config.hidden_size - rope_theta = getattr(config, "rope_theta", 10000) - rope_scaling = getattr(config, "rope_scaling", None) - max_position_embeddings = getattr(config, "max_position_embeddings", - 8192) - self.self_attn = XverseAttention( - hidden_size=self.hidden_size, - num_heads=config.num_attention_heads, - num_kv_heads=getattr(config, "num_key_value_heads", - config.num_attention_heads), - rope_theta=rope_theta, - rope_scaling=rope_scaling, - max_position_embeddings=max_position_embeddings, - quant_config=quant_config, - bias=getattr(config, "bias", False), - cache_config=cache_config, - prefix=f"{prefix}.self_attn", - ) - self.mlp = XverseMLP( - hidden_size=self.hidden_size, - intermediate_size=config.intermediate_size, - hidden_act=config.hidden_act, - quant_config=quant_config, - ) - self.input_layernorm = RMSNorm(config.hidden_size, - eps=config.rms_norm_eps) - self.post_attention_layernorm = RMSNorm(config.hidden_size, - eps=config.rms_norm_eps) - - def forward( - self, - positions: torch.Tensor, - hidden_states: torch.Tensor, - kv_cache: torch.Tensor, - attn_metadata: AttentionMetadata, - residual: Optional[torch.Tensor], - ) -> Tuple[torch.Tensor, torch.Tensor]: - # Self Attention - if residual is None: - residual = hidden_states - hidden_states = self.input_layernorm(hidden_states) - else: - hidden_states, residual = self.input_layernorm( - hidden_states, residual) - hidden_states = self.self_attn( - positions=positions, - hidden_states=hidden_states, - kv_cache=kv_cache, - attn_metadata=attn_metadata, - ) - - # Fully Connected - hidden_states, residual = self.post_attention_layernorm( - hidden_states, residual) - hidden_states = self.mlp(hidden_states) - return hidden_states, residual - - -@support_torch_compile -class XverseModel(nn.Module): - - def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""): - super().__init__() - config = vllm_config.model_config.hf_config - cache_config = vllm_config.cache_config - quant_config = vllm_config.quant_config - lora_config = vllm_config.lora_config - self.config = config - self.padding_idx = config.pad_token_id - lora_vocab = (lora_config.lora_extra_vocab_size * - (lora_config.max_loras or 1)) if lora_config else 0 - self.vocab_size = config.vocab_size + lora_vocab - self.org_vocab_size = config.vocab_size - self.embed_tokens = VocabParallelEmbedding( - self.vocab_size, - config.hidden_size, - org_num_embeddings=config.vocab_size, - ) - self.start_layer, self.end_layer, self.layers = make_layers( - config.num_hidden_layers, - lambda prefix: XverseDecoderLayer( - config, cache_config, quant_config, prefix=prefix), - 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 get_input_embeddings(self, input_ids: torch.Tensor) -> torch.Tensor: - return self.embed_tokens(input_ids) - - def forward( - self, - input_ids: torch.Tensor, - positions: torch.Tensor, - kv_caches: List[torch.Tensor], - attn_metadata: AttentionMetadata, - intermediate_tensors: Optional[IntermediateTensors], - inputs_embeds: Optional[torch.Tensor] = None, - ) -> Union[torch.Tensor, IntermediateTensors]: - if get_pp_group().is_first_rank: - if inputs_embeds is not None: - hidden_states = inputs_embeds - else: - hidden_states = self.get_input_embeddings(input_ids) - residual = None - else: - 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 - 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, residual) - return hidden_states - - -class XverseForCausalLM(nn.Module, SupportsLoRA, SupportsPP): - packed_modules_mapping = { - "qkv_proj": [ - "q_proj", - "k_proj", - "v_proj", - ], - "gate_up_proj": [ - "gate_proj", - "up_proj", - ], - } - - # LoRA specific attributes - supported_lora_modules = [ - "qkv_proj", - "o_proj", - "gate_up_proj", - "down_proj", - "embed_tokens", - "lm_head", - ] - embedding_modules = { - "embed_tokens": "input_embeddings", - "lm_head": "output_embeddings", - } - embedding_padding_modules = ["lm_head"] - - def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""): - super().__init__() - - config = vllm_config.model_config.hf_config - quant_config = vllm_config.quant_config - lora_config = vllm_config.lora_config - - self.config = config - self.lora_config = lora_config - - self.quant_config = quant_config - self.model = XverseModel(vllm_config=vllm_config, - prefix=maybe_prefix(prefix, "model")) - self.lm_head = ParallelLMHead(config.vocab_size, - config.hidden_size, - quant_config=quant_config) - if self.config.tie_word_embeddings: - self.lm_head.weight = self.model.embed_tokens.weight - self.logits_processor = LogitsProcessor(config.vocab_size) - self.sampler = get_sampler() - self.make_empty_intermediate_tensors = ( - self.model.make_empty_intermediate_tensors) - - def get_input_embeddings(self, input_ids: torch.Tensor) -> torch.Tensor: - return self.model.get_input_embeddings(input_ids) - - def forward( - self, - input_ids: torch.Tensor, - positions: torch.Tensor, - kv_caches: List[torch.Tensor], - attn_metadata: AttentionMetadata, - intermediate_tensors: Optional[IntermediateTensors] = None, - inputs_embeds: Optional[torch.Tensor] = None, - ) -> Union[torch.Tensor, IntermediateTensors]: - hidden_states = self.model(input_ids, positions, kv_caches, - attn_metadata, intermediate_tensors, - inputs_embeds) - return hidden_states - - def compute_logits( - self, - hidden_states: torch.Tensor, - sampling_metadata: SamplingMetadata, - ) -> Optional[torch.Tensor]: - logits = self.logits_processor(self.lm_head, hidden_states, - sampling_metadata) - return logits - - def sample( - self, - logits: torch.Tensor, - sampling_metadata: SamplingMetadata, - ) -> Optional[SamplerOutput]: - next_tokens = self.sampler(logits, sampling_metadata) - return next_tokens - - def load_weights(self, weights: Iterable[Tuple[str, - torch.Tensor]]) -> Set[str]: - stacked_params_mapping = [ - ("qkv_proj", "q_proj", "q"), - ("qkv_proj", "k_proj", "k"), - ("qkv_proj", "v_proj", "v"), - ("gate_up_proj", "gate_proj", 0), - ("gate_up_proj", "up_proj", 1), - ] - params_dict = dict(self.named_parameters()) - loaded_params: Set[str] = set() - for name, loaded_weight in weights: - if ("rotary_emb.inv_freq" in name - or "rotary_emb.cos_cached" in name - or "rotary_emb.sin_cached" in name): - continue - for (param_name, weight_name, shard_id) in stacked_params_mapping: - if weight_name not in name: - continue - name = name.replace(weight_name, param_name) - # 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) - break - else: - # 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) - weight_loader(param, loaded_weight) - loaded_params.add(name) - return loaded_params