Spaces:
Running
on
Zero
Running
on
Zero
File size: 5,348 Bytes
c62903f |
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 |
from typing import List, Optional, Tuple, Union
import torch
import torch.nn as nn
import transformers
from transformers import (
AutoConfig,
AutoModelForCausalLM,
Qwen2Config,
Qwen2ForCausalLM,
Qwen2Model,
)
from transformers.generation.utils import GenerateOutput
from transformers.modeling_outputs import CausalLMOutputWithPast
from ..egogpt_arch import EgoGPTMetaForCausalLM, EgoGPTMetaModel
class EgoGPTConfigQwen(Qwen2Config):
model_type = "egogpt_qwen"
class EgoGPTQwenModel(EgoGPTMetaModel, Qwen2Model):
config_class = EgoGPTConfigQwen
def __init__(self, config: Qwen2Config):
super(EgoGPTQwenModel, self).__init__(config)
class EgoGPTQwenForCausalLM(Qwen2ForCausalLM, EgoGPTMetaForCausalLM):
config_class = EgoGPTConfigQwen
def __init__(self, config):
super(Qwen2ForCausalLM, self).__init__(config)
config.rope_scaling = None
self.model = EgoGPTQwenModel(config)
self.vocab_size = config.vocab_size
self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
# Initialize weights and apply final processing
self.post_init()
def get_model(self):
return self.model
def forward(
self,
input_ids: torch.LongTensor = None,
attention_mask: Optional[torch.Tensor] = None,
position_ids: Optional[torch.LongTensor] = None,
past_key_values: Optional[List[torch.FloatTensor]] = None,
inputs_embeds: Optional[torch.FloatTensor] = None,
labels: Optional[torch.LongTensor] = None,
use_cache: Optional[bool] = None,
output_attentions: Optional[bool] = None,
output_hidden_states: Optional[bool] = None,
speech: Optional[torch.FloatTensor] = None,
speech_lengths: Optional[torch.LongTensor] = None,
images: Optional[torch.FloatTensor] = None,
image_sizes: Optional[List[List[int]]] = None,
modalities: Optional[List[str]] = ["image"],
return_dict: Optional[bool] = None,
cache_position: Optional[torch.LongTensor] = None,
) -> Union[Tuple, CausalLMOutputWithPast]:
if inputs_embeds is None:
(
input_ids,
position_ids,
attention_mask,
past_key_values,
inputs_embeds,
labels,
) = self.prepare_inputs_labels_for_speech_and_text(
input_ids,
position_ids,
attention_mask,
past_key_values,
labels,
speech,
speech_lengths,
images,
image_sizes,
modalities,
)
return super().forward(
input_ids=input_ids,
attention_mask=attention_mask,
position_ids=position_ids,
past_key_values=past_key_values,
inputs_embeds=inputs_embeds,
labels=labels,
use_cache=use_cache,
output_attentions=output_attentions,
output_hidden_states=output_hidden_states,
return_dict=return_dict,
)
@torch.no_grad()
def generate(
self,
inputs: Optional[torch.Tensor] = None,
speech: Optional[torch.Tensor] = None,
speech_lengths: Optional[torch.Tensor] = None,
images: Optional[torch.FloatTensor] = None,
image_sizes: Optional[List[List[int]]] = None,
modalities: Optional[List[str]] = ["image"],
**kwargs,
) -> Union[GenerateOutput, torch.LongTensor]:
position_ids = kwargs.pop("position_ids", None)
attention_mask = kwargs.pop("attention_mask", None)
if "inputs_embeds" in kwargs:
raise NotImplementedError("`inputs_embeds` is not supported")
if speech is not None:
(
inputs,
position_ids,
attention_mask,
_,
inputs_embeds,
_,
) = self.prepare_inputs_labels_for_speech_and_text(
inputs,
position_ids,
attention_mask,
None,
None,
speech,
speech_lengths,
images,
image_sizes,
modalities,
)
else:
inputs_embeds = self.get_model().embed_tokens(inputs)
return super().generate(
position_ids=position_ids,
attention_mask=attention_mask,
inputs_embeds=inputs_embeds,
**kwargs,
)
def prepare_inputs_for_generation(
self, input_ids, past_key_values=None, inputs_embeds=None, **kwargs
):
speech = kwargs.pop("speech", None)
speech_lengths = kwargs.pop("speech_lengths", None)
inputs = super().prepare_inputs_for_generation(
input_ids,
past_key_values=past_key_values,
inputs_embeds=inputs_embeds,
**kwargs,
)
if speech is not None:
inputs["speech"] = speech
inputs["speech_lengths"] = speech_lengths
return inputs
AutoConfig.register("egogpt_qwen", EgoGPTConfigQwen)
AutoModelForCausalLM.register(EgoGPTConfigQwen, EgoGPTQwenForCausalLM)
|