File size: 7,537 Bytes
5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 2010c83 5169b80 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 |
from dataclasses import fields
from typing import List, Optional, Tuple, Union
import torch
import torch.nn.functional as F
import math
from transformers import PreTrainedModel
from transformers.modeling_outputs import CausalLMOutputWithPast, BaseModelOutputWithPast
from transformers.models.auto import AutoModelForCausalLM
from .config import ModelConfig
from .model import OLMo
from .configuration_olmo import OLMoConfig
def create_model_config_from_pretrained_config(config: OLMoConfig):
"""
Utility function
"""
kwargs = {}
for field in fields(ModelConfig):
kwargs[field.name] = getattr(config, field.name)
model_config = ModelConfig(**kwargs)
return model_config
class OLMoPreTrainedModel(PreTrainedModel):
config_class = OLMoConfig
base_model_prefix = "model"
_no_split_modules = ["OLMoBlock"]
# _skip_keys_device_placement = ["past_key_values", "causal_mask"]
_skip_keys_device_placement = ["past_key_values"]
def _init_weights(self, module):
# `OLMoModel.reset_parameters` initializes weights of itself and its children
if isinstance(module, OLMo):
module.reset_parameters()
class OLMoForCausalLM(OLMoPreTrainedModel):
_tied_weights_keys = []
# _tied_weights_keys = ["transformer.wte.weight"]
def __init__(self, config: OLMoConfig):
super().__init__(config)
self.model = OLMo(config)
# Initialize weights and apply final processing
self.post_init()
def get_input_embeddings(self) -> torch.nn.Module:
return self.model.transformer.wte
def set_input_embeddings(self, value: torch.nn.Module):
self.model.transformer.wte = value
def get_output_embeddings(self):
if self.config.weight_tying:
return self.model.transformer.wte
else:
return self.model.transformer.ff_out
def set_output_embeddings(self, value: torch.nn.Module):
if self.config.weight_tying:
self.model.transformer.wte = value
else:
self.model.transformer.ff_out = value
def set_decoder(self, decoder):
self.model = decoder
def get_decoder(self):
return self.model
def forward(
self,
input_ids: torch.LongTensor = None,
inputs_embeds: Optional[torch.FloatTensor] = None,
attention_mask: Optional[torch.Tensor] = None,
attention_bias: Optional[torch.Tensor] = None,
past_key_values: Optional[List[torch.FloatTensor]] = None,
labels: Optional[torch.LongTensor] = None,
use_cache: Optional[bool] = None,
output_attentions: Optional[bool] = None,
output_hidden_states: Optional[bool] = None,
return_dict: Optional[bool] = None,
) -> Union[Tuple, CausalLMOutputWithPast]:
r"""
Args:
labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
(masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
Returns:
Example:
```python
>>> from transformers import AutoTokenizer, OLMoForCausalLM
>>> model = OLMoForCausalLM.from_pretrained("allenai/OLMo-7B")
>>> tokenizer = AutoTokenizer.from_pretrained("allenai/OLMo-7B")
>>> prompt = "Hey, are you conscious? Can you talk to me?"
>>> inputs = tokenizer(prompt, return_tensors="pt")
>>> # Generate
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
```"""
output_attentions = output_attentions or self.config.output_attentions
output_hidden_states = output_hidden_states or self.config.output_hidden_states
use_cache = use_cache if use_cache is not None else self.config.use_cache
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
assert not output_attentions
# decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
base_output: Union[BaseModelOutputWithPast, Tuple] = self.model.forward(
input_ids=input_ids,
inputs_embeds=inputs_embeds,
attention_mask=attention_mask,
attention_bias=attention_bias,
past_key_values=past_key_values,
use_cache=use_cache,
output_hidden_states=output_hidden_states,
)
last_hidden_state = base_output.last_hidden_state if return_dict else base_output[0]
# Get logits.
# shape: (batch_size, seq_len or 1, vocab_size)
if self.config.weight_tying:
logits = F.linear(last_hidden_state, self.model.transformer.wte.weight, None) # type: ignore
else:
logits = self.model.transformer.ff_out(last_hidden_state) # type: ignore
if self.config.scale_logits:
logits.mul_(1 / math.sqrt(self.config.d_model))
loss = None
if labels is not None:
# Shift so that tokens < n predict n
shift_logits = logits[..., :-1, :].contiguous()
shift_labels = labels[..., 1:].contiguous()
# Flatten the tokens
loss_fct = torch.nn.CrossEntropyLoss()
shift_logits = shift_logits.view(-1, self.config.vocab_size)
shift_labels = shift_labels.view(-1)
# Enable model parallelism
shift_labels = shift_labels.to(shift_logits.device)
loss = loss_fct(shift_logits, shift_labels)
if not return_dict:
output = (logits,) + base_output[1:]
return (loss,) + output if loss is not None else output
assert isinstance(base_output, BaseModelOutputWithPast)
return CausalLMOutputWithPast(
loss=loss,
logits=logits,
past_key_values=base_output.past_key_values,
hidden_states=base_output.hidden_states,
attentions=base_output.attentions,
)
def prepare_inputs_for_generation(
self, input_ids: torch.LongTensor, past_key_values: Optional[List[Tuple]] = None, **kwargs
):
if past_key_values:
# This is because we want the model to only process the last generated token.
input_ids = input_ids[:, -1:]
model_inputs = {"input_ids": input_ids, "past_key_values": past_key_values}
kwargs.pop("cache_position")
model_inputs.update(kwargs)
# logger.warning("%s %s", kwargs.keys(), model_inputs.keys())
# model_inputs["use_cache"] = kwargs.pop("use_cache", self.config.use_cache)
return model_inputs
@staticmethod
def _reorder_cache(past_key_values, beam_idx):
reordered_past = ()
for layer_past in past_key_values:
reordered_past += (
tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past),
)
return reordered_past
# Register the model so that it is available for transformer pipelines, auto-loading, etc.
AutoModelForCausalLM.register(OLMoConfig, OLMoForCausalLM)
|