update causal batch for ct2 and fix nf4 (#17)

* update causal batch for ct2 and fix nf4

* bump the ctranslate2 version

---------

Co-authored-by: Michael Feil <michael.feil@michaelfeil.eu>
This commit is contained in:
Michael Feil 2024-02-09 11:07:14 -08:00 committed by GitHub
parent 339ede9e90
commit 972e9a7f7c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 457 additions and 17 deletions

View File

@ -53,7 +53,7 @@ You may set the `TGICHAT_(USER|ASS|SYS)_(PRE|POST)` environment variables, to wr
```bash
model=TheBloke/Llama-2-13B-Chat-fp16 # around 14GB Vram.
volume=$PWD/data # share a volume with the Docker container to avoid downloading weights every run
image=docker.io/michaelf34/tgi:03-10-2023 # docker image by @michaelfeil
image=docker.io/michaelf34/tgi:05-11-2023 # docker image by @michaelfeil
docker run --gpus all --shm-size 1g -p 8080:80 -v $volume:/data $image --model-id $model --quantize ct2
```

View File

@ -15,9 +15,11 @@ grpcio-status = "^1.51.1"
grpcio-reflection = "^1.51.1"
grpc-interceptor = "^0.15.0"
typer = "^0.6.1"
accelerate = { version = "^0.19.0", optional = true }
ctranslate2 = { version = "^3.20.0", optional = true }
bitsandbytes = { version = "^0.40.0", optional = true }
accelerate = { version = "^0.20.3", optional = true }
ctranslate2 = { version = "^3.23.0", optional = true }
bitsandbytes = { version = "^0.41.1", optional = true }
torch = { version = "^2.0.1" }
scipy = "^1.11.3"
safetensors = "0.3.1"
loguru = "^0.6.0"
opentelemetry-api = "^1.15.0"
@ -26,8 +28,8 @@ opentelemetry-instrumentation-grpc = "^0.36b0"
hf-transfer = "^0.1.2"
sentencepiece = "^0.1.97"
tokenizers = "0.13.3"
huggingface-hub = "^0.14.1"
transformers = "4.29.2"
huggingface-hub = "^0.15.1"
transformers = "4.32.1"
einops = "^0.6.1"
texttable = { version = "^1.6.7", optional = true }
datasets = { version = "^2.14.0", optional = true }

View File

@ -1,10 +1,10 @@
accelerate==0.19.0 ; python_version >= "3.9" and python_version < "4.0"
accelerate==0.20.3 ; python_version >= "3.9" and python_version < "4.0"
aiohttp==3.8.5 ; python_version >= "3.9" and python_version < "4.0"
aiosignal==1.3.1 ; python_version >= "3.9" and python_version < "4.0"
async-timeout==4.0.2 ; python_version >= "3.9" and python_version < "4.0"
attrs==23.1.0 ; python_version >= "3.9" and python_version < "4.0"
backoff==2.2.1 ; python_version >= "3.9" and python_version < "4.0"
bitsandbytes==0.38.1 ; python_version >= "3.9" and python_version < "4.0"
bitsandbytes==0.41.1 ; python_version >= "3.9" and python_version < "4.0"
certifi==2023.5.7 ; python_version >= "3.9" and python_version < "4.0"
charset-normalizer==3.1.0 ; python_version >= "3.9" and python_version < "4.0"
click==8.1.3 ; python_version >= "3.9" and python_version < "4.0"
@ -23,7 +23,7 @@ grpcio-reflection==1.56.0 ; python_version >= "3.9" and python_version < "4.0"
grpcio-status==1.56.0 ; python_version >= "3.9" and python_version < "4.0"
grpcio==1.56.0 ; python_version >= "3.9" and python_version < "4.0"
hf-transfer==0.1.3 ; python_version >= "3.9" and python_version < "4.0"
huggingface-hub==0.14.1 ; python_version >= "3.9" and python_version < "4.0"
huggingface-hub==0.15.1 ; python_version >= "3.9" and python_version < "4.0"
idna==3.4 ; python_version >= "3.9" and python_version < "4.0"
jinja2==3.1.2 ; python_version >= "3.9" and python_version < "4.0"
loguru==0.6.0 ; python_version >= "3.9" and python_version < "4.0"
@ -56,12 +56,13 @@ safetensors==0.3.1 ; python_version >= "3.9" and python_version < "4.0"
sentencepiece==0.1.99 ; python_version >= "3.9" and python_version < "4.0"
setuptools==68.0.0 ; python_version >= "3.9" and python_version < "4.0"
six==1.16.0 ; python_version >= "3.9" and python_version < "4.0"
scipy==1.11.3 ; python_version >= "3.9" and python_version < "4.0"
sympy==1.12 ; python_version >= "3.9" and python_version < "4.0"
texttable==1.6.7 ; python_version >= "3.9" and python_version < "4.0"
tokenizers==0.13.3 ; python_version >= "3.9" and python_version < "4.0"
torch==2.0.1 ; python_version >= "3.9" and python_version < "4.0"
tqdm==4.65.0 ; python_version >= "3.9" and python_version < "4.0"
transformers==4.29.2 ; python_version >= "3.9" and python_version < "4.0"
transformers==4.32.1 ; python_version >= "3.9" and python_version < "4.0"
typer==0.6.1 ; python_version >= "3.9" and python_version < "4.0"
typing-extensions==4.7.1 ; python_version >= "3.9" and python_version < "4.0"
tzdata==2023.3 ; python_version >= "3.9" and python_version < "4.0"

View File

@ -23,12 +23,20 @@ import numpy as np
import os
import multiprocessing
from pathlib import Path
from dataclasses import dataclass
from huggingface_hub.constants import HUGGINGFACE_HUB_CACHE
from opentelemetry import trace
from transformers import (
AutoTokenizer,
AutoConfig,
PreTrainedTokenizerBase
)
from text_generation_server.models.types import (
Batch,
PrefillTokens,
Generation,
GeneratedText,
)
from typing import Optional, Tuple, List, Type, Dict
@ -38,9 +46,10 @@ from text_generation_server.models.types import (
Generation,
GeneratedText,
)
from text_generation_server.pb import generate_pb2
from text_generation_server.utils import NextTokenChooser, StoppingCriteria, Sampling
from text_generation_server.utils import Sampling
from text_generation_server.models.causal_lm import CausalLMBatch
try:
import ctranslate2
@ -51,6 +60,434 @@ except ImportError:
tracer = trace.get_tracer(__name__)
@dataclass
class CT2CausalLMBatch(Batch):
batch_id: int
requests: List[generate_pb2.Request]
requests_idx_mapping: Dict[int, int]
# Decoder values
input_ids: torch.Tensor
attention_mask: torch.Tensor
position_ids: torch.Tensor
past_key_values: Optional[List[Tuple]]
# All tokens
all_input_ids: List[torch.Tensor]
# Lengths of all generations present in the batch
input_lengths: List[int]
prefix_offsets: List[int]
read_offsets: List[int]
# Generation helpers
next_token_choosers: List[NextTokenChooser]
stopping_criterias: List[StoppingCriteria]
# Metadata used for padding
max_input_length: int
padding_right_offset: int
# Maximum number of tokens this batch will grow to
max_tokens: int
# Past metadata
keys_head_dim_last: bool = True
def to_pb(self) -> generate_pb2.CachedBatch:
return generate_pb2.CachedBatch(
id=self.batch_id,
request_ids=[r.id for r in self.requests],
size=len(self),
max_tokens=self.max_tokens,
)
@classmethod
def from_pb(
cls,
pb: generate_pb2.Batch,
tokenizer: PreTrainedTokenizerBase,
dtype: torch.dtype,
device: torch.device,
) -> "CT2CausalLMBatch":
inputs = []
next_token_choosers = []
stopping_criterias = []
prefix_offsets = []
read_offsets = []
requests_idx_mapping = {}
# Parse batch
max_truncation = 0
padding_right_offset = 0
max_decode_tokens = 0
for i, r in enumerate(pb.requests):
requests_idx_mapping[r.id] = i
inputs.append(r.inputs)
next_token_choosers.append(NextTokenChooser.from_pb(r.parameters, device))
stopping_criteria = StoppingCriteria.from_pb(
r.stopping_parameters, tokenizer
)
stopping_criterias.append(stopping_criteria)
max_truncation = max(max_truncation, r.truncate)
max_decode_tokens += stopping_criteria.max_new_tokens
padding_right_offset = max(
padding_right_offset, stopping_criteria.max_new_tokens
)
tokenized_inputs = tokenizer(
inputs,
return_tensors="pt",
padding=True,
return_token_type_ids=False,
truncation=True,
max_length=max_truncation,
).to(device)
for _ in pb.requests:
input_len = tokenized_inputs["input_ids"].shape[1]
prefix_offsets.append(input_len - 5)
read_offsets.append(input_len)
input_lengths = tokenized_inputs["attention_mask"].sum(1)
max_input_length = input_lengths.max()
input_ids = tokenized_inputs["input_ids"]
# Allocate maximum attention_mask
attention_mask = input_ids.new_zeros(
(pb.size, max_input_length + padding_right_offset)
)
# Copy tokenizer attention_mask into fully allocated attention_mask
attention_mask[:, :max_input_length] = tokenized_inputs["attention_mask"]
position_ids = tokenized_inputs["attention_mask"].long().cumsum(-1) - 1
position_ids.masked_fill_(tokenized_inputs["attention_mask"] == 0, 1)
all_input_ids = tokenized_inputs["input_ids"].T.split(1, dim=1)
max_tokens = len(inputs) * (max_input_length + max_decode_tokens)
return cls(
batch_id=pb.id,
requests=pb.requests,
requests_idx_mapping=requests_idx_mapping,
input_ids=input_ids,
attention_mask=attention_mask,
position_ids=position_ids,
past_key_values=None,
all_input_ids=list(all_input_ids),
input_lengths=input_lengths.tolist(),
prefix_offsets=prefix_offsets,
read_offsets=read_offsets,
next_token_choosers=next_token_choosers,
stopping_criterias=stopping_criterias,
max_input_length=max_input_length.item(),
padding_right_offset=padding_right_offset,
max_tokens=max_tokens,
)
@tracer.start_as_current_span("filter")
def filter(self, request_ids: List[int]) -> Optional["CT2CausalLMBatch"]:
if len(request_ids) == 0:
raise ValueError("Batch must have at least one request")
if len(request_ids) == len(self):
return self
keep_indices = []
# New values after filtering
requests_idx_mapping = {}
requests = []
input_lengths = []
prefix_offsets = []
read_offsets = []
all_input_ids = []
max_input_length = 0
next_token_choosers = []
stopping_criterias = []
total_remaining_decode_tokens = 0
new_padding_right_offset = 0
for i, request_id in enumerate(request_ids):
idx = self.requests_idx_mapping[request_id]
requests_idx_mapping[request_id] = i
keep_indices.append(idx)
requests.append(self.requests[idx])
prefix_offsets.append(self.prefix_offsets[idx])
read_offsets.append(self.read_offsets[idx])
all_input_ids.append(self.all_input_ids[idx])
request_input_length = self.input_lengths[idx]
input_lengths.append(request_input_length)
max_input_length = max(max_input_length, request_input_length)
next_token_choosers.append(self.next_token_choosers[idx])
stopping_criteria = self.stopping_criterias[idx]
stopping_criterias.append(stopping_criteria)
remaining_decode_tokens = (
stopping_criteria.max_new_tokens - stopping_criteria.current_tokens
)
total_remaining_decode_tokens += remaining_decode_tokens
new_padding_right_offset = max(
new_padding_right_offset, remaining_decode_tokens
)
# Apply indices to input_ids, attention mask, past key values and other items that need to be cached
input_ids = self.input_ids[keep_indices]
position_ids = self.position_ids[keep_indices]
self.attention_mask = self.attention_mask[
keep_indices,
-(self.padding_right_offset + max_input_length) : (
self.attention_mask.shape[1] - self.padding_right_offset
)
+ new_padding_right_offset,
]
# Ensure that past_key_values tensors can be updated in-place
if type(self.past_key_values[0]) == tuple:
self.past_key_values = [list(layer) for layer in self.past_key_values]
# Update tensors in-place to allow incremental garbage collection
past_kv_length = max_input_length - 1
for layer in self.past_key_values:
past_keys, past_values = layer
if len(past_keys.shape) == 3:
# Force past to be of dim [self_size, num_heads, ...] for easy indexing
past_keys = past_keys.view(len(self), -1, *past_keys.shape[-2:])
past_values = past_values.view(len(self), -1, *past_values.shape[-2:])
if self.keys_head_dim_last:
layer[0] = past_keys[keep_indices, :, -past_kv_length:, :]
else:
layer[0] = past_keys[keep_indices, :, :, -past_kv_length:]
del past_keys
layer[1] = past_values[keep_indices, :, -past_kv_length:, :]
del past_values
max_tokens = len(request_ids) * max_input_length + total_remaining_decode_tokens
self.requests = requests
self.requests_idx_mapping = requests_idx_mapping
self.input_ids = input_ids
self.position_ids = position_ids
self.all_input_ids = all_input_ids
self.input_lengths = input_lengths
self.prefix_offsets = prefix_offsets
self.read_offsets = read_offsets
self.next_token_choosers = next_token_choosers
self.stopping_criterias = stopping_criterias
self.max_input_length = max_input_length
self.padding_right_offset = new_padding_right_offset
self.max_tokens = max_tokens
return self
@classmethod
@tracer.start_as_current_span("concatenate")
def concatenate(cls, batches: List["CT2CausalLMBatch"]) -> "CT2CausalLMBatch":
# Used for padding
total_batch_size = 0
max_input_length = 0
padding_right_offset = 0
for batch in batches:
total_batch_size += len(batch)
max_input_length = max(max_input_length, batch.max_input_length)
padding_right_offset = max(padding_right_offset, batch.padding_right_offset)
# Batch attributes
requests = []
requests_idx_mapping = {}
input_lengths = []
prefix_offsets = []
read_offsets = []
all_input_ids = []
next_token_choosers = []
stopping_criterias = []
max_tokens = 0
# Batch tensors
input_ids = None
attention_mask = None
position_ids = None
past_key_values = []
# Used for slicing correctly inside the tensors
# Equivalent to a cumsum on batch sizes
start_index = 0
for i, batch in enumerate(batches):
requests.extend(batch.requests)
input_lengths.extend(batch.input_lengths)
prefix_offsets.extend(batch.prefix_offsets)
read_offsets.extend(batch.read_offsets)
all_input_ids.extend(batch.all_input_ids)
next_token_choosers.extend(batch.next_token_choosers)
stopping_criterias.extend(batch.stopping_criterias)
if i == 0:
requests_idx_mapping = batch.requests_idx_mapping
else:
# We need to offset the mapping for each batch by the cumulative batch size
for k, v in batch.requests_idx_mapping.items():
requests_idx_mapping[k] = v + start_index
# Slicing end index for this batch
end_index = start_index + len(batch)
# We only concatenate batches that did at least one step
# if batch.past_key_values is None:
# raise ValueError("only concatenate prefilled batches")
# Create empty tensor
# input_ids is always of shape [batch_size, 1]
# We do not need to pad it
if input_ids is None:
input_ids = batch.input_ids.new_empty((total_batch_size, 1))
# Copy to correct indices
input_ids[start_index:end_index] = batch.input_ids
# Create padded tensor
if attention_mask is None:
attention_mask = batch.attention_mask.new_zeros(
(total_batch_size, max_input_length + padding_right_offset),
)
# We need to slice the attention mask to remove padding from previous steps
# and to remove unused allocated space
left_offset = max_input_length - batch.max_input_length
batch_left_offset = (
batch.attention_mask.shape[1]
- batch.max_input_length
- batch.padding_right_offset
)
attention_mask[
start_index:end_index,
left_offset:-padding_right_offset,
] = batch.attention_mask[
:,
batch_left_offset : -batch.padding_right_offset,
]
# Create empty tensor
# position_ids is always of shape [batch_size, 1]
if position_ids is None:
position_ids = batch.position_ids.new_empty((total_batch_size, 1))
position_ids[start_index:end_index] = batch.position_ids
# Shenanigans to get dimensions because BLOOM outputs a past with a different shape
# BLOOM Keys: [batch_size * num_heads, head_dim, seq_length]
# BLOOM Values: [batch_size * num_heads, seq_length, head_dim]
# And ensure that we can update tensors in-place
# if type(batch.past_key_values[0]) == tuple:
# batch.past_key_values = [
# [t.view(len(batch), -1, *t.shape[-2:]) for t in layer]
# for layer in batch.past_key_values
# ]
# elif len(batch.past_key_values[0][0].shape) == 3:
# for layer in batch.past_key_values:
# for k, t in enumerate(layer):
# layer[k] = t.view(len(batch), -1, *t.shape[-2:])
# Add eventual padding tokens that were added while concatenating
max_tokens += batch.max_tokens + (
max_input_length - batch.max_input_length
) * len(batch)
start_index = end_index
# first_past_kvs = batches[0].past_key_values
# _, num_heads, padded_sequence_length, head_dim = first_past_kvs[0][1].shape
# padded_past_values_shape = (
# total_batch_size,
# num_heads,
# max_input_length - 1,
# head_dim,
# )
# if batches[0].keys_head_dim_last:
# padded_past_keys_shape = padded_past_values_shape
# else:
# # seq_length is last for BLOOM
# padded_past_keys_shape = (
# total_batch_size,
# num_heads,
# head_dim,
# max_input_length - 1,
# )
# Iterate over attention layers
# Concatenate past key values layer by layer to allow incremental garbage collection
# for j in range(len(first_past_kvs)):
# padded_past_keys = first_past_kvs[j][0].new_zeros(padded_past_keys_shape)
# start_index = 0
# for batch in batches:
# past_keys = batch.past_key_values[j][0]
# # Clear reference to the original tensor
# batch.past_key_values[j][0] = None
# # Slicing end index for this batch
# end_index = start_index + len(batch)
# # We slice the keys to remove the padding from previous batches
# past_seq_len = batch.max_input_length - 1
# if batch.keys_head_dim_last:
# padded_past_keys[
# start_index:end_index, :, -past_seq_len:, :
# ] = past_keys[:, :, -past_seq_len:, :]
# else:
# # BLOOM case
# padded_past_keys[
# start_index:end_index, :, :, -past_seq_len:
# ] = past_keys[:, :, :, -past_seq_len:]
# del past_keys
# start_index = end_index
# padded_past_values = first_past_kvs[j][1].new_zeros(
# padded_past_values_shape
# )
# start_index = 0
# for batch in batches:
# past_values = batch.past_key_values[j][1]
# # Clear reference to the original tensor
# batch.past_key_values[j][1] = None
# # Slicing end index for this batch
# end_index = start_index + len(batch)
# # We slice the past values to remove the padding from previous batches
# past_seq_len = batch.max_input_length - 1
# padded_past_values[
# start_index:end_index, :, -past_seq_len:, :
# ] = past_values[:, :, -past_seq_len:, :]
# del past_values
# # Update values
# start_index = end_index
# past_key_values.append([padded_past_keys, padded_past_values])
return cls(
batch_id=batches[0].batch_id,
requests=requests,
requests_idx_mapping=requests_idx_mapping,
input_ids=input_ids,
attention_mask=attention_mask,
position_ids=position_ids,
past_key_values=past_key_values,
all_input_ids=all_input_ids,
input_lengths=input_lengths,
prefix_offsets=prefix_offsets,
read_offsets=read_offsets,
next_token_choosers=next_token_choosers,
stopping_criterias=stopping_criterias,
max_input_length=max_input_length,
padding_right_offset=padding_right_offset,
keys_head_dim_last=batches[0].keys_head_dim_last,
max_tokens=max_tokens,
)
def __len__(self):
return len(self.requests)
class CT2CausalLM(Model):
def __init__(
self,
@ -176,8 +613,8 @@ class CT2CausalLM(Model):
)
@property
def batch_type(self) -> Type[CausalLMBatch]:
return CausalLMBatch
def batch_type(self) -> Type[CT2CausalLMBatch]:
return CT2CausalLMBatch
def decode(self, generated_ids: List[int]) -> str:
return self.tokenizer.decode(
@ -221,8 +658,8 @@ class CT2CausalLM(Model):
@tracer.start_as_current_span("generate_token")
def generate_token(
self, batch: CausalLMBatch
) -> Tuple[List[Generation], Optional[CausalLMBatch]]:
self, batch: CT2CausalLMBatch
) -> Tuple[List[Generation], Optional[CT2CausalLMBatch]]:
logits, past = self.forward_ct2(batch.all_input_ids, batch.input_lengths)
# Results

View File

@ -185,8 +185,8 @@ class FlashLlamaAttention(torch.nn.Module):
self.hidden_size = config.hidden_size
self.head_size = self.hidden_size // self.num_heads
self.rotary_emb = PositionRotaryEmbedding.load(
prefix=f"{prefix}.rotary_emb", weights=weights
self.rotary_emb = PositionRotaryEmbedding.static(
dim=self.head_size, device=weights.device, base=10000.0,
)
self.softmax_scale = self.head_size**-0.5