deeplotx 0.5.5__tar.gz → 0.5.6__tar.gz

{deeplotx-0.5.5 → deeplotx-0.5.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deeplotx
-Version: 0.5.5
+Version: 0.5.6
 Summary: Easy-2-use long text NLP toolkit.
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
@@ -265,7 +265,8 @@ Dynamic: license-file
     long_text_encoder = LongTextEncoder(
         max_length=2048,  # 最大文本大小, 超出截断
         chunk_size=448,  # 块大小 (按 Token 计)
-        overlapping=32  # 块间重叠大小 (按 Token 计)
+        overlapping=32,  # 块间重叠大小 (按 Token 计)
+        cache_capacity=512  # 缓存大小
     )
 
     trainer = TextBinaryClassifierTrainer(

{deeplotx-0.5.5 → deeplotx-0.5.6}/README.md

@@ -247,7 +247,8 @@
     long_text_encoder = LongTextEncoder(
         max_length=2048,  # 最大文本大小, 超出截断
         chunk_size=448,  # 块大小 (按 Token 计)
-        overlapping=32  # 块间重叠大小 (按 Token 计)
+        overlapping=32,  # 块间重叠大小 (按 Token 计)
+        cache_capacity=512  # 缓存大小
     )
 
     trainer = TextBinaryClassifierTrainer(

{deeplotx-0.5.5 → deeplotx-0.5.6}/deeplotx/encoder/encoder.py

@@ -25,7 +25,7 @@ class Encoder(nn.Module):
         self.embed_dim = self.encoder.config.max_position_embeddings
         logger.debug(f'{Encoder.__name__} initialized on device: {self.device}.')
 
-    def forward(self, input_ids: torch.Tensor, attention_mask: torch.Tensor) -> torch.Tensor:
+    def forward(self, input_ids: torch.Tensor, attention_mask: torch.Tensor, *args, **kwargs) -> torch.Tensor:
         def _encoder(_input_tup: tuple[torch.Tensor, torch.Tensor]) -> torch.Tensor:
             return self.encoder.forward(_input_tup[0], attention_mask=_input_tup[1]).last_hidden_state[:, 0, :]
 

{deeplotx-0.5.5 → deeplotx-0.5.6}/deeplotx/encoder/long_text_encoder.py

@@ -25,6 +25,10 @@ class LongTextEncoder(Encoder):
     def __chunk_embedding(self, idx: int, x: torch.Tensor, mask: torch.Tensor) -> tuple[int, torch.Tensor]:
         return idx, super().forward(x, attention_mask=mask)
 
+    @override
+    def forward(self, text: str, flatten: bool = False, *args, **kwargs) -> torch.Tensor:
+        return self.encode(text=text, flatten=flatten)
+
     @override
     def encode(self, text: str, flatten: bool = False) -> torch.Tensor:
         def postprocess(tensors: list[torch.Tensor], _flatten: bool) -> torch.Tensor:

{deeplotx-0.5.5 → deeplotx-0.5.6}/deeplotx/trainer/text_binary_classification_trainer.py

@@ -31,7 +31,7 @@ class TextBinaryClassifierTrainer(BaseTrainer):
             positive_texts = positive_texts[:min_length]
             negative_texts = negative_texts[:min_length]
         all_texts = positive_texts + negative_texts
-        text_embeddings = [self._long_text_encoder.encode(x, flatten=False, use_cache=True) for x in all_texts]
+        text_embeddings = [self._long_text_encoder.encode(x, flatten=False) for x in all_texts]
         feature_dim = text_embeddings[0].shape[-1]
         dtype = text_embeddings[0].dtype
         labels = ([torch.tensor([1.], dtype=dtype, device=self.device) for _ in range(len(positive_texts))]

{deeplotx-0.5.5 → deeplotx-0.5.6}/deeplotx.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deeplotx
-Version: 0.5.5
+Version: 0.5.6
 Summary: Easy-2-use long text NLP toolkit.
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
@@ -265,7 +265,8 @@ Dynamic: license-file
     long_text_encoder = LongTextEncoder(
         max_length=2048,  # 最大文本大小, 超出截断
         chunk_size=448,  # 块大小 (按 Token 计)
-        overlapping=32  # 块间重叠大小 (按 Token 计)
+        overlapping=32,  # 块间重叠大小 (按 Token 计)
+        cache_capacity=512  # 缓存大小
     )
 
     trainer = TextBinaryClassifierTrainer(

{deeplotx-0.5.5 → deeplotx-0.5.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "deeplotx"
-version = "0.5.5"
+version = "0.5.6"
 description = "Easy-2-use long text NLP toolkit."
 readme = "README.md"
 requires-python = ">=3.10"