feat(strategy): 重构mask构建逻辑并优化策略工厂参数传递

khaosz/trainer/mask.py

@@ -1,55 +0,0 @@
-
-import torch
-from abc import abstractmethod
-from torch import Tensor
-
-
-
-class MaskBuilder:
-    def __init__(
-        self,
-        bos_token_id: int,
-        eos_token_id: int,
-        user_token_id: int,
-        system_token_id: int,
-        
-    ):
-        self.bos_token_id = bos_token_id
-        self.eos_token_id = eos_token_id
-        self.user_token_id = user_token_id
-        self.system_token_id = system_token_id
-    
-    @abstractmethod
-    def build(input_ids: Tensor) -> Tensor:
-        raise NotImplementedError
-
-
-
-class LossMaskBuilder(MaskBuilder):
-    def __init__(self, **kwargs):
-        super().__init__(**kwargs)
-    
-    def build(self, input_ids: Tensor) -> Tensor:
-        token_markers = torch.zeros_like(input_ids, dtype=torch.int8)
-        
-        is_user_token = input_ids.eq(self.user_token_id)
-        is_system_token = input_ids.eq(self.system_token_id)
-        
-        token_markers[is_user_token] = 1
-        token_markers[is_system_token] = -1 
-        
-        cumulative_markers = torch.cumsum(token_markers, dim=-1)
-        min_cumulative = cumulative_markers.min(dim=-1, keepdim=True).values
-        loss_mask = cumulative_markers - min_cumulative
-    
-        return loss_mask
-        
-        
-        
-
-class AttentionMaskBuilder:
-    def __init__(self, **kwargs):
-        super().__init__(**kwargs)
-    
-    def build(input_ids: Tensor):
-        bsz = input_ids.size(0)

khaosz/trainer/strategy.py

@@ -32,67 +32,34 @@ def get_logprobs(model:nn.Module, input_ids: Tensor, mask: Tensor, pad_token_id)
     
     return (token_logprobs * valid_mask).sum(dim=-1)
 
+def build_loss_mask(input_ids: Tensor, bos_token_id: int, eos_token_id: int) -> Tensor:
+    token_markers = torch.zeros_like(input_ids, dtype=torch.int8)
     
-class MaskBuilder:
+    is_bos_token = input_ids.eq(bos_token_id)
-    def __init__(
+    is_eos_token = input_ids.eq(eos_token_id)
-        self,
-        bos_token_id: int,
-        eos_token_id: int,
-        user_token_id: int,
-        system_token_id: int,
     
-    ):
+    token_markers[is_bos_token] = 1
-        self.bos_token_id = bos_token_id
+    token_markers[is_eos_token] = -1 
-        self.eos_token_id = eos_token_id
-        self.user_token_id = user_token_id
-        self.system_token_id = system_token_id
     
-    @abstractmethod
+    cumulative_markers = torch.cumsum(token_markers, dim=-1)
-    def build(input_ids: Tensor) -> Tensor:
+    min_cumulative = cumulative_markers.min(dim=-1, keepdim=True).values
-        raise NotImplementedError
+    loss_mask = cumulative_markers - min_cumulative
 
+    return loss_mask.to(dtype=torch.bool)
 
+def build_attention_mask(input_ids: Tensor, user_token_id: int, multi_turn: bool = False) -> Tensor:
+    bsz, seq_len = input_ids.size()
+    is_user_token = input_ids.eq(user_token_id)
+    turn_id = is_user_token.cumsum(dim=-1)
     
-class LossMaskBuilder(MaskBuilder):
+    iq = turn_id.view(bsz, seq_len, 1)
-    def __init__(self, **kwargs):
+    ik = turn_id.view(bsz, 1, seq_len)
-        super().__init__(**kwargs)
-    
-    def build(self, input_ids: Tensor) -> Tensor:
-        token_markers = torch.zeros_like(input_ids, dtype=torch.int8)
-        
-        is_user_token = input_ids.eq(self.user_token_id)
-        is_system_token = input_ids.eq(self.system_token_id)
-        
-        token_markers[is_user_token] = 1
-        token_markers[is_system_token] = -1 
-        
-        cumulative_markers = torch.cumsum(token_markers, dim=-1)
-        min_cumulative = cumulative_markers.min(dim=-1, keepdim=True).values
-        loss_mask = cumulative_markers - min_cumulative
-    
-        return loss_mask.to(dtype=torch.bool)
-        
-        
-        
-
-class AttentionMaskBuilder(MaskBuilder):
-    def __init__(self, multi_turn=False, **kwargs):
-        super().__init__(**kwargs)
-        self.multi_turn = multi_turn
-    
-    def build(self, input_ids: Tensor):
-        bsz = input_ids.size(0)
-        
-    
-    def _build_batch(self, input_ids: Tensor):
-        is_user_token = input_ids.eq(self.user_token_id)
-        
-        token_markers = torch.zeros_like(input_ids, dtype=torch.int8)
-        token_markers[is_user_token] = 1
-        cumulative_markers = torch.cumsum(token_markers, dim=-1)
-            
     
+    seq_mask = (iq <= ik) if multi_turn else (iq == ik)
+    causal_mask = torch.tril(torch.ones(seq_len, seq_len, device=input_ids.device)).bool()
+    attention_mask = seq_mask & causal_mask
     
+    return attention_mask
 
 
 class BaseStrategy(ABC):
@@ -209,11 +176,11 @@ class PpoStrategy(BaseStrategy):
 
 class StrategyFactory:
     
-    def load(model, train_type, pad_token_id, dpo_beta):
+    def load(model, train_type, **kwargs):
         train_strategy: Dict[str, Callable[[], BaseStrategy]] = {
             "seq": lambda: SeqStrategy(model),
             "sft": lambda: SftStrategy(model),
-            "dpo": lambda: DpoStrategy(model, pad_token_id, dpo_beta)
+            "dpo": lambda: DpoStrategy(model, kwargs.pop("pad_token_id") , kwargs.pop("dpo_beta"))
         }
         strategy = train_strategy[train_type]()
         return strategy