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

Re-enable the 80 char line width limit (#3305)

Browse Source
This commit is contained in:
Zhuohan Li
2024-03-10 19:49:14 -07:00
committed by GitHub
parent 4b59f00e91
commit 2f8844ba08
67 changed files with 557 additions and 528 deletions
Download Patch File Download Diff File Expand all files Collapse all files

89
vllm/transformers_utils/configs/mpt.py
View File

@@ -62,62 +62,6 @@ class MPTConfig(PretrainedConfig):
fc_type: str = 'torch',
verbose: Optional[int] = None,
**kwargs: Any):
"""The MPT configuration class.
Args:
d_model (int): The size of the embedding dimension of the model.
n_heads (int): The number of attention heads.
n_layers (int): The number of layers in the model.
expansion_ratio (int): The ratio of the up/down scale in the ffn.
max_seq_len (int): The maximum sequence length of the model.
vocab_size (int): The size of the vocabulary.
resid_pdrop (float): The dropout probability applied to the attention output before combining with residual.
emb_pdrop (float): The dropout probability for the embedding layer.
learned_pos_emb (bool): Whether to use learned positional embeddings
attn_config (Dict): A dictionary used to configure the model's attention module:
attn_type (str): type of attention to use. Options: multihead_attention, multiquery_attention, grouped_query_attention
attn_pdrop (float): The dropout probability for the attention layers.
attn_impl (str): The attention implementation to use. One of 'torch', 'flash', or 'triton'.
qk_ln (bool): Whether to apply layer normalization to the queries and keys in the attention layer.
clip_qkv (Optional[float]): If not None, clip the queries, keys, and values in the attention layer to
this value.
softmax_scale (Optional[float]): If not None, scale the softmax in the attention layer by this value. If None,
use the default scale of ``1/sqrt(d_keys)``.
prefix_lm (Optional[bool]): Whether the model should operate as a Prefix LM. This requires passing an
extra `prefix_mask` argument which indicates which tokens belong to the prefix. Tokens in the prefix
can attend to one another bi-directionally. Tokens outside the prefix use causal attention.
attn_uses_sequence_id (Optional[bool]): Whether to restrict attention to tokens that have the same sequence_id.
When the model is in `train` mode, this requires passing an extra `sequence_id` argument which indicates
which sub-sequence each token belongs to.
Defaults to ``False`` meaning any provided `sequence_id` will be ignored.
alibi (bool): Whether to use the alibi bias instead of position embeddings.
alibi_bias_max (int): The maximum value of the alibi bias.
kv_n_heads (Optional[int]): For grouped_query_attention only, allow user to specify number of kv heads.
ffn_config (Dict): A dictionary used to configure the model's ffn module:
ffn_type (str): type of ffn to use. Options: mptmlp, te_ln_mlp
init_device (str): The device to use for parameter initialization.
logit_scale (Optional[Union[float, str]]): If not None, scale the logits by this value.
no_bias (bool): Whether to use bias in all layers.
verbose (int): The verbosity level. 0 is silent.
embedding_fraction (float): The fraction to scale the gradients of the embedding layer by.
norm_type (str): choose type of norm to use
use_cache (bool): Whether or not the model should return the last key/values attentions
init_config (Dict): A dictionary used to configure the model initialization:
init_config.name: The parameter initialization scheme to use. Options: 'default_', 'baseline_',
'kaiming_uniform_', 'kaiming_normal_', 'neox_init_', 'small_init_', 'xavier_uniform_', or
'xavier_normal_'. These mimic the parameter initialization methods in PyTorch.
init_div_is_residual (Union[int, float, str, bool]): Value to divide initial weights by if ``module._is_residual`` is True.
emb_init_std (Optional[float]): The standard deviation of the normal distribution used to initialize the embedding layer.
emb_init_uniform_lim (Optional[Union[Tuple[float, float], float]]): The lower and upper limits of the uniform distribution
used to initialize the embedding layer. Mutually exclusive with ``emb_init_std``.
init_std (float): The standard deviation of the normal distribution used to initialize the model,
if using the baseline_ parameter initialization scheme.
init_gain (float): The gain to use for parameter initialization with kaiming or xavier initialization schemes.
fan_mode (str): The fan mode to use for parameter initialization with kaiming initialization schemes.
init_nonlinearity (str): The nonlinearity to use for parameter initialization with kaiming initialization schemes.
---
See llmfoundry.models.utils.param_init_fns.py for info on other param init config options
fc_type (str): choose fc layer implementation. Options: torch and te. te layers support fp8 when using H100 GPUs.
"""
self.d_model = d_model
self.n_heads = n_heads
self.n_layers = n_layers
@@ -139,8 +83,8 @@ class MPTConfig(PretrainedConfig):
self.fc_type = fc_type
if verbose is not None:
warnings.warn(DeprecationWarning(
'verbose argument for MPTConfig is now ignored and will be removed. Use python_log_level instead.'
),
'verbose argument for MPTConfig is now ignored and '
'will be removed. Use python_log_level instead.'),
stacklevel=2)
if 'name' in kwargs:
del kwargs['name']
@@ -149,7 +93,8 @@ class MPTConfig(PretrainedConfig):
if self.attn_config.get('alibi', False):
self.learned_pos_emb = False
warnings.warn(
f'alibi is turned on, setting `learned_pos_emb` to {self.learned_pos_emb}`',
f'alibi is turned on, setting `learned_pos_emb` '
f'to {self.learned_pos_emb}`',
stacklevel=2)
super().__init__(**kwargs)
self._validate_config()
@@ -176,8 +121,8 @@ class MPTConfig(PretrainedConfig):
[self.attn_config['attn_pdrop'], self.resid_pdrop, self.emb_pdrop]
)):
raise ValueError(
"self.attn_config['attn_pdrop'], resid_pdrop, emb_pdrop are probabilities and must be between 0 and 1" # pylint: disable=line-too-long
)
"self.attn_config['attn_pdrop'], resid_pdrop, emb_pdrop are "
"probabilities and must be between 0 and 1")
if self.attn_config['attn_impl'] not in ['torch', 'flash', 'triton']:
raise ValueError(
f"Unknown attn_impl={self.attn_config['attn_impl']}")
@@ -193,17 +138,17 @@ class MPTConfig(PretrainedConfig):
if self.attn_config['attn_uses_sequence_id'] and self.attn_config[
'attn_impl'] not in ['torch', 'triton']:
raise NotImplementedError(
'attn_uses_sequence_id only implemented with torch and triton attention.' # pylint: disable=line-too-long
)
'attn_uses_sequence_id only implemented with torch '
'and triton attention.')
if self.embedding_fraction > 1 or self.embedding_fraction <= 0:
raise ValueError(
'model.embedding_fraction must be between 0 (exclusive) and 1 (inclusive)!' # pylint: disable=line-too-long
)
'model.embedding_fraction must be between 0 (exclusive) '
'and 1 (inclusive)!')
if isinstance(self.logit_scale,
str) and self.logit_scale != 'inv_sqrt_d_model':
raise ValueError(
f"self.logit_scale={self.logit_scale!r} is not recognized as an option; use numeric value or 'inv_sqrt_d_model'." # pylint: disable=line-too-long
)
f"self.logit_scale={self.logit_scale!r} is not recognized as "
"an option; use numeric value or 'inv_sqrt_d_model'.")
if self.init_config.get('name', None) is None:
raise ValueError(
f"self.init_config={self.init_config!r} 'name' needs to be set."
@@ -219,11 +164,11 @@ class MPTConfig(PretrainedConfig):
del te
except Exception as exc:
raise ImportError(
# pylint: disable=line-too-long
'TransformerEngine import fail. `fc_type: te` requires TransformerEngine be installed. '
+
'The required version of transformer_engine also requires FlashAttention v1.0.6 is installed:\n'
+ 'pip install flash-attn==1.0.6 --no-build-isolation \n' +
'TransformerEngine import fail. `fc_type: te` requires '
'TransformerEngine be installed. '
'The required version of transformer_engine also requires '
'FlashAttention v1.0.6 is installed:\n'
'pip install flash-attn==1.0.6 --no-build-isolation \n'
'pip install git+https://github.com/NVIDIA/TransformerEngine.git@144e4888b2cdd60bd52e706d5b7a79cb9c1a7156'
) from exc
if self.ffn_config['ffn_type'] == 'mptmlp':

72
vllm/transformers_utils/configs/starcoder2.py
View File

@@ -2,78 +2,6 @@ from transformers import PretrainedConfig
class Starcoder2Config(PretrainedConfig):
r"""
This is the configuration class to store the configuration of a [`Starcoder2Model`]. It is used to instantiate a
Starcoder2 model according to the specified arguments, defining the model architecture. Instantiating a configuration
with the defaults will yield a similar configuration to that of the [bigcode/starcoder2-7b_16k](https://huggingface.co/bigcode/starcoder2-7b_16k) model.
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.
Args:
vocab_size (`int`, *optional*, defaults to 49152):
Vocabulary size of the Starcoder2 model. Defines the number of different tokens that can be represented by the
`inputs_ids` passed when calling [`Starcoder2Model`]
hidden_size (`int`, *optional*, defaults to 3072):
Dimension of the hidden representations.
intermediate_size (`int`, *optional*, defaults to 12288):
Dimension of the MLP representations.
num_hidden_layers (`int`, *optional*, defaults to 30):
Number of hidden layers in the Transformer encoder.
num_attention_heads (`int`, *optional*, defaults to 24):
Number of attention heads for each attention layer in the Transformer encoder.
num_key_value_heads (`int`, *optional*, defaults to 2):
This is the number of key_value heads that should be used to implement Grouped Query Attention. If
`num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
`num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When
converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
by meanpooling all the original heads within that group. For more details checkout [this
paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to `8`.
hidden_act (`str` or `function`, *optional*, defaults to `"gelu_pytorch_tanh"`):
The non-linear activation function (function or string) in the decoder.
max_position_embeddings (`int`, *optional*, defaults to 4096):
The maximum sequence length that this model might ever be used with. Starcoder2's sliding window attention
allows sequence of up to 4096*32 tokens.
initializer_range (`float`, *optional*, defaults to 0.02):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
norm_epsilon (`float`, *optional*, defaults to 1e-05):
Epsilon value for the layer norm
use_cache (`bool`, *optional*, defaults to `True`):
Whether or not the model should return the last key/values attentions (not used by all models). Only
relevant if `config.is_decoder=True`.
bos_token_id (`int`, *optional*, defaults to 50256):
The id of the "beginning-of-sequence" token.
eos_token_id (`int`, *optional*, defaults to 50256):
The id of the "end-of-sequence" token.
rope_theta (`float`, *optional*, defaults to 10000.0):
The base period of the RoPE embeddings.
sliding_window (`int`, *optional*):
Sliding window attention window size. If not specified, will default to `None` (no sliding window).
attention_dropout (`float`, *optional*, defaults to 0.0):
The dropout ratio for the attention probabilities.
residual_dropout (`float`, *optional*, defaults to 0.0):
Residual connection dropout value.
embedding_dropout (`float`, *optional*, defaults to 0.0):
Embedding dropout.
use_bias (`bool`, *optional*, defaults to `True`):
Whether to use bias term on linear layers of the model.
```python
>>> from transformers import Starcoder2Model, Starcoder2Config
>>> # Initializing a Starcoder2 7B style configuration
>>> configuration = Starcoder2Config()
>>> # Initializing a model from the Starcoder2 7B style configuration
>>> model = Starcoder2Model(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
```"""
model_type = "starcoder2"
keys_to_ignore_at_inference = ["past_key_values"]