Re-enable the 80 char line width limit (#3305)
This commit is contained in:
@@ -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':
|
||||
|
||||
@@ -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"]
|
||||
|
||||
|
||||
@@ -1,4 +1,3 @@
|
||||
# yapf: disable
|
||||
# Adapted from
|
||||
# https://huggingface.co/baichuan-inc/Baichuan2-13B-Chat/blob/8f6e343d545c503b91429582231d1d354dac2740/tokenization_baichuan.py
|
||||
# This includes a fix suggested in
|
||||
@@ -13,7 +12,6 @@ import sentencepiece as spm
|
||||
from transformers.tokenization_utils import AddedToken, PreTrainedTokenizer
|
||||
from transformers.utils import logging
|
||||
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
VOCAB_FILES_NAMES = {"vocab_file": "tokenizer.model"}
|
||||
@@ -52,27 +50,16 @@ class BaichuanTokenizer(PreTrainedTokenizer):
|
||||
clean_up_tokenization_spaces=False,
|
||||
**kwargs,
|
||||
):
|
||||
self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
|
||||
bos_token = (
|
||||
AddedToken(bos_token, lstrip=False, rstrip=False)
|
||||
if isinstance(bos_token, str)
|
||||
else bos_token
|
||||
)
|
||||
eos_token = (
|
||||
AddedToken(eos_token, lstrip=False, rstrip=False)
|
||||
if isinstance(eos_token, str)
|
||||
else eos_token
|
||||
)
|
||||
unk_token = (
|
||||
AddedToken(unk_token, lstrip=False, rstrip=False)
|
||||
if isinstance(unk_token, str)
|
||||
else unk_token
|
||||
)
|
||||
pad_token = (
|
||||
AddedToken(pad_token, lstrip=False, rstrip=False)
|
||||
if isinstance(pad_token, str)
|
||||
else pad_token
|
||||
)
|
||||
self.sp_model_kwargs = ({} if sp_model_kwargs is None else
|
||||
sp_model_kwargs)
|
||||
bos_token = (AddedToken(bos_token, lstrip=False, rstrip=False)
|
||||
if isinstance(bos_token, str) else bos_token)
|
||||
eos_token = (AddedToken(eos_token, lstrip=False, rstrip=False)
|
||||
if isinstance(eos_token, str) else eos_token)
|
||||
unk_token = (AddedToken(unk_token, lstrip=False, rstrip=False)
|
||||
if isinstance(unk_token, str) else unk_token)
|
||||
pad_token = (AddedToken(pad_token, lstrip=False, rstrip=False)
|
||||
if isinstance(pad_token, str) else pad_token)
|
||||
self.vocab_file = vocab_file
|
||||
self.add_bos_token = add_bos_token
|
||||
self.add_eos_token = add_eos_token
|
||||
@@ -107,7 +94,10 @@ class BaichuanTokenizer(PreTrainedTokenizer):
|
||||
|
||||
def get_vocab(self):
|
||||
"""Returns vocab as a dict"""
|
||||
vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
|
||||
vocab = {
|
||||
self.convert_ids_to_tokens(i): i
|
||||
for i in range(self.vocab_size)
|
||||
}
|
||||
vocab.update(self.added_tokens_encoder)
|
||||
return vocab
|
||||
|
||||
@@ -130,7 +120,8 @@ class BaichuanTokenizer(PreTrainedTokenizer):
|
||||
out_string = ""
|
||||
prev_is_special = False
|
||||
for i, token in enumerate(tokens):
|
||||
# make sure that special tokens are not decoded using sentencepiece model
|
||||
# make sure that special tokens are not decoded using
|
||||
# sentencepiece model
|
||||
if token in self.all_special_tokens:
|
||||
if not prev_is_special and i != 0:
|
||||
out_string += " "
|
||||
@@ -143,9 +134,9 @@ class BaichuanTokenizer(PreTrainedTokenizer):
|
||||
out_string += self.sp_model.decode(current_sub_tokens)
|
||||
return out_string
|
||||
|
||||
def save_vocabulary(
|
||||
self, save_directory, filename_prefix: Optional[str] = None
|
||||
) -> Tuple[str]:
|
||||
def save_vocabulary(self,
|
||||
save_directory,
|
||||
filename_prefix: Optional[str] = None) -> Tuple[str]:
|
||||
"""
|
||||
Save the vocabulary and special tokens file to a directory.
|
||||
|
||||
@@ -157,24 +148,24 @@ class BaichuanTokenizer(PreTrainedTokenizer):
|
||||
`Tuple(str)`: Paths to the files saved.
|
||||
"""
|
||||
if not os.path.isdir(save_directory):
|
||||
logger.error(f"Vocabulary path ({save_directory}) should be a directory")
|
||||
logger.error(f"Vocabulary path ({save_directory}) "
|
||||
"should be a directory")
|
||||
return
|
||||
out_vocab_file = os.path.join(
|
||||
save_directory,
|
||||
(filename_prefix + "-" if filename_prefix else "")
|
||||
+ VOCAB_FILES_NAMES["vocab_file"],
|
||||
(filename_prefix + "-" if filename_prefix else "") +
|
||||
VOCAB_FILES_NAMES["vocab_file"],
|
||||
)
|
||||
|
||||
if os.path.abspath(self.vocab_file) != os.path.abspath(
|
||||
out_vocab_file
|
||||
) and os.path.isfile(self.vocab_file):
|
||||
out_vocab_file) and os.path.isfile(self.vocab_file):
|
||||
copyfile(self.vocab_file, out_vocab_file)
|
||||
elif not os.path.isfile(self.vocab_file):
|
||||
with open(out_vocab_file, "wb") as fi:
|
||||
content_spiece_model = self.sp_model.serialized_model_proto()
|
||||
fi.write(content_spiece_model)
|
||||
|
||||
return (out_vocab_file,)
|
||||
return (out_vocab_file, )
|
||||
|
||||
def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
|
||||
bos_token_id = [self.bos_token_id] if self.add_bos_token else []
|
||||
@@ -194,7 +185,8 @@ class BaichuanTokenizer(PreTrainedTokenizer):
|
||||
already_has_special_tokens: bool = False,
|
||||
) -> List[int]:
|
||||
"""
|
||||
Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
|
||||
Retrieve sequence ids from a token list that has no special tokens
|
||||
added. This method is called when adding
|
||||
special tokens using the tokenizer `prepare_for_model` method.
|
||||
|
||||
Args:
|
||||
@@ -202,11 +194,14 @@ class BaichuanTokenizer(PreTrainedTokenizer):
|
||||
List of IDs.
|
||||
token_ids_1 (`List[int]`, *optional*):
|
||||
Optional second list of IDs for sequence pairs.
|
||||
already_has_special_tokens (`bool`, *optional*, defaults to `False`):
|
||||
Whether or not the token list is already formatted with special tokens for the model.
|
||||
already_has_special_tokens (`bool`, *optional*, defaults to
|
||||
`False`):
|
||||
Whether or not the token list is already formatted with
|
||||
special tokens for the model.
|
||||
|
||||
Returns:
|
||||
`List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
|
||||
`List[int]`: A list of integers in the range [0, 1]:
|
||||
1 for a special token, 0 for a sequence token.
|
||||
"""
|
||||
if already_has_special_tokens:
|
||||
return super().get_special_tokens_mask(
|
||||
@@ -220,20 +215,16 @@ class BaichuanTokenizer(PreTrainedTokenizer):
|
||||
|
||||
if token_ids_1 is None:
|
||||
return bos_token_id + ([0] * len(token_ids_0)) + eos_token_id
|
||||
return (
|
||||
bos_token_id
|
||||
+ ([0] * len(token_ids_0))
|
||||
+ eos_token_id
|
||||
+ bos_token_id
|
||||
+ ([0] * len(token_ids_1))
|
||||
+ eos_token_id
|
||||
)
|
||||
return (bos_token_id + ([0] * len(token_ids_0)) + eos_token_id +
|
||||
bos_token_id + ([0] * len(token_ids_1)) + eos_token_id)
|
||||
|
||||
def create_token_type_ids_from_sequences(
|
||||
self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
|
||||
) -> List[int]:
|
||||
self,
|
||||
token_ids_0: List[int],
|
||||
token_ids_1: Optional[List[int]] = None) -> List[int]:
|
||||
"""
|
||||
Creates a mask from the two sequences passed to be used in a sequence-pair classification task. An ALBERT
|
||||
Creates a mask from the two sequences passed to be used in a
|
||||
sequence-pair classification task. An ALBERT
|
||||
sequence pair mask has the following format:
|
||||
|
||||
```
|
||||
@@ -250,7 +241,8 @@ class BaichuanTokenizer(PreTrainedTokenizer):
|
||||
Optional second list of IDs for sequence pairs.
|
||||
|
||||
Returns:
|
||||
`List[int]`: List of [token type IDs](../glossary#token-type-ids) according to the given sequence(s).
|
||||
`List[int]`: List of [token type IDs](../glossary#token-type-ids)
|
||||
according to the given sequence(s).
|
||||
"""
|
||||
bos_token_id = [self.bos_token_id] if self.add_bos_token else []
|
||||
eos_token_id = [self.eos_token_id] if self.add_eos_token else []
|
||||
|
||||
Reference in New Issue
Block a user