|
import torch |
|
import torch.nn as nn |
|
import re |
|
|
|
from transformers import BertConfig, BertModel |
|
|
|
|
|
class IdentityMap(nn.Module): |
|
def __init__(self): |
|
super().__init__() |
|
|
|
def forward(self, x, *args, **kwargs): |
|
return x |
|
|
|
@property |
|
def config(self): |
|
return {"mm_projector_type": 'identity'} |
|
|
|
|
|
class SimpleResBlock(nn.Module): |
|
def __init__(self, channels): |
|
super().__init__() |
|
self.pre_norm = nn.LayerNorm(channels) |
|
|
|
self.proj = nn.Sequential( |
|
nn.Linear(channels, channels), |
|
nn.GELU(), |
|
nn.Linear(channels, channels) |
|
) |
|
def forward(self, x): |
|
x = self.pre_norm(x) |
|
return x + self.proj(x) |
|
|
|
|
|
def build_vision_projector(config, delay_load=False, **kwargs): |
|
projector_type = getattr(config, 'mm_projector_type', 'linear') |
|
|
|
if projector_type == 'linear': |
|
return nn.Linear(config.mm_hidden_size, config.hidden_size) |
|
|
|
mlp_gelu_match = re.match(r'^mlp(\d+)x_gelu$', projector_type) |
|
if mlp_gelu_match: |
|
mlp_depth = int(mlp_gelu_match.group(1)) |
|
modules = [nn.Linear(config.mm_hidden_size, config.hidden_size)] |
|
for _ in range(1, mlp_depth): |
|
modules.append(nn.GELU()) |
|
modules.append(nn.Linear(config.hidden_size, config.hidden_size)) |
|
return nn.Sequential(*modules) |
|
|
|
if projector_type == 'identity': |
|
return IdentityMap() |
|
|
|
raise ValueError(f'Unknown projector type: {projector_type}') |
|
|
|
class ImageEmbeddingPooler(nn.Module): |
|
|
|
def __init__(self): |
|
super().__init__() |
|
self.embedding_dim = 512 |
|
|
|
|
|
config = BertConfig( |
|
hidden_size=self.embedding_dim, |
|
num_hidden_layers=2, |
|
num_attention_heads=8, |
|
intermediate_size=self.embedding_dim*4, |
|
use_position_embeddings=True, |
|
max_position_embeddings=2304, |
|
use_bfloat16=True, |
|
vocab_size=1, |
|
) |
|
self.bert = BertModel(config) |
|
|
|
def forward(self, embeddings, attention_mask): |
|
|
|
batch_size, num_tokens, _ = embeddings.shape |
|
|
|
outputs = self.bert(inputs_embeds=embeddings, attention_mask=attention_mask) |
|
|
|
outputs = outputs['last_hidden_state'].to(embeddings.dtype)[:, :196] |
|
|
|
return outputs |
|
|
|
def build_image_pooler(config): |
|
return ImageEmbeddingPooler() |
|
|