File size: 6,297 Bytes
326e148 2625b05 326e148 2625b05 326e148 54ed2c9 326e148 2625b05 326e148 2625b05 326e148 2625b05 326e148 2625b05 326e148 2625b05 326e148 2625b05 326e148 2625b05 326e148 2625b05 326e148 2625b05 | 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 | import torch
import torch.nn as nn
from transformers import PreTrainedModel, AutoConfig, AutoModel
from .configuration_emcoder import EmCoderConfig
class EmCoderEncoder(nn.Module):
"""The core encoder architecture of EmCoder Transformer."""
def __init__(self, config: EmCoderConfig):
super().__init__()
self.token_embedding = nn.Embedding(config.vocab_size, config.d_model)
self.pos_embedding = nn.Embedding(config.max_seq_len, config.d_model)
self.embed_norm = nn.LayerNorm(config.d_model)
encoder_layer = nn.TransformerEncoderLayer(
d_model=config.d_model,
nhead=config.n_head,
dim_feedforward=config.d_ffn,
dropout=config.dropout,
activation="gelu",
norm_first=True,
batch_first=True,
)
self.encoder = nn.TransformerEncoder(
encoder_layer=encoder_layer, num_layers=config.n_layers, enable_nested_tensor=False
)
self.final_norm = nn.LayerNorm(config.d_model)
self.dropout = nn.Dropout(config.dropout)
def forward(self, x: torch.Tensor, mask: torch.Tensor) -> torch.Tensor:
"""Standard forward pass through the encoder."""
seq_len = x.size(1)
pos_ids = torch.arange(seq_len, device=x.device).unsqueeze(0)
x = self.token_embedding(x) + self.pos_embedding(pos_ids)
x = self.embed_norm(x)
x = self.dropout(x)
padding_mask = mask == 0
encoded = self.encoder(x, src_key_padding_mask=padding_mask)
return self.final_norm(encoded)
class EmCoder(PreTrainedModel):
"""The full EmCoder model, including the classification head."""
config_class = EmCoderConfig
def __init__(self, config: EmCoderConfig):
super().__init__(config)
self.encoder = EmCoderEncoder(config)
self.classifier = nn.Sequential(
nn.Linear(config.d_model, config.d_model),
nn.GELU(),
nn.Dropout(config.dropout),
nn.Linear(config.d_model, config.num_labels),
)
self.post_init()
def _init_weights(self, module: nn.Module) -> None:
if isinstance(module, nn.Linear):
nn.init.trunc_normal_(module.weight, std=0.02)
if module.bias is not None:
nn.init.zeros_(module.bias)
elif isinstance(module, nn.Embedding):
nn.init.trunc_normal_(module.weight, std=0.02)
if hasattr(module, "padding_idx") and module.padding_idx is not None:
module.weight.data[module.padding_idx].zero_()
elif isinstance(module, nn.LayerNorm):
nn.init.ones_(module.weight)
nn.init.zeros_(module.bias)
def _set_mc_dropout(self, active: bool = True):
for m in self.modules():
if isinstance(m, nn.Dropout) or isinstance(m, nn.MultiheadAttention):
m.train(active)
@staticmethod
def _masked_mean_pooling(
features: torch.Tensor, mask: torch.Tensor
) -> torch.Tensor:
mask = mask.unsqueeze(-1) # (B, S, 1)
masked_features = features * mask # (B, S, D)
sum_masked_features = masked_features.sum(dim=1) # (B, D)
count_tokens = torch.clamp(mask.sum(dim=1), min=1e-9) # (B, 1)
return sum_masked_features / count_tokens # (B, D)
def mc_forward(
self,
input_ids: torch.Tensor | None = None,
attention_mask: torch.Tensor | None = None,
n_samples: int = 10,
max_batch_size: int | None = None,
return_dict: bool | None = None,
**kwargs,
) -> torch.Tensor:
"""
Performs Monte Carlo Dropout inference to quantify epistemic uncertainty.
Args:
x: Input token IDs of shape (B, S).
mask: Attention mask of shape (B, S).
n_samples: Total number of Monte Carlo samples.
max_batch_size: Maximum number of samples in one forward pass.
Returns:
Logits of shape (n_samples, B, num_labels).
"""
x = input_ids if input_ids is not None else kwargs.get("x")
mask = attention_mask if attention_mask is not None else kwargs.get("mask")
if x is None or mask is None:
raise ValueError("input_ids (x) and attention_mask (mask) must be provided")
if max_batch_size is None:
max_batch_size = n_samples
B, S = x.shape
num_labels = self.classifier[-1].out_features
all_logits = torch.empty((n_samples, B, num_labels), device=x.device)
is_training = self.training
self._set_mc_dropout(active=True)
try:
for i in range(0, n_samples, max_batch_size):
batch_samples = min(max_batch_size, n_samples - i)
x_stacked = x.repeat(batch_samples, 1) # (batch_samples * B, S)
mask_stacked = mask.repeat(batch_samples, 1) # (batch_samples * B, S)
features = self.encoder(
x_stacked, mask_stacked
) # (batch_samples * B, S, D)
pooled = self._masked_mean_pooling(features, mask_stacked)
logits = self.classifier(pooled) # (n_samples * B, num_labels)
all_logits[i : i + batch_samples] = logits.view(batch_samples, B, -1)
finally:
self._set_mc_dropout(active=is_training)
return all_logits
def forward(
self,
input_ids: torch.Tensor | None = None,
attention_mask: torch.Tensor | None = None,
return_dict: bool | None = None,
**kwargs,
) -> torch.Tensor:
"""Standard forward pass without MC Dropout."""
x = input_ids if input_ids is not None else kwargs.get("x")
mask = attention_mask if attention_mask is not None else kwargs.get("mask")
if x is None or mask is None:
raise ValueError("input_ids (x) and attention_mask (mask) must be provided")
features = self.encoder(x, mask)
pooled = self._masked_mean_pooling(features, mask)
return self.classifier(pooled)
try:
AutoConfig.register("emcoder", EmCoderConfig)
AutoModel.register(EmCoderConfig, EmCoder)
except ValueError:
pass |