deeplotx 0.2.18__py3-none-any.whl → 0.2.20__py3-none-any.whl

deeplotx/nn/base_neural_network.py

@@ -23,10 +23,10 @@ class BaseNeuralNetwork(nn.Module):
             for param in self.parameters():
                 l2_reg += (torch.pow(param, exponent=2.)).sum()
             return l2_reg
-        return _lambda * _l2()
+        return _lambda * _l2() / 2.
 
     def elastic_net(self, alpha: float = 1e-4, rho: float = 0.5) -> torch.Tensor:
-        return alpha * rho * self.l1(_lambda=1.) + alpha * (1 - rho) * self.l2(_lambda=1.) / 2.
+        return alpha * (rho * self.l1(_lambda=1.) + (1 - rho) * self.l2(_lambda=1.))
 
     def forward(self, x) -> torch.Tensor: ...
 
@@ -39,9 +39,9 @@ class BaseNeuralNetwork(nn.Module):
         return res
 
     def save(self):
-        torch.save(self.state_dict(), f'{self._model_name}.deeplotx.pth')
+        torch.save(self.state_dict(), f'{self._model_name}.deeplotx')
         return self
 
     def load(self):
-        self.load_state_dict(torch.load(f'{self._model_name}.deeplotx.pth'))
+        self.load_state_dict(torch.load(f'{self._model_name}.deeplotx'))
         return self

deeplotx/trainer/text_binary_classification_trainer.py

@@ -21,8 +21,7 @@ class TextBinaryClassifierTrainer(BaseTrainer):
     def train(self, positive_texts: list[str], negative_texts: list[str],
               num_epochs: int, learning_rate: float = 2e-5, balancing_dataset: bool = True,
               train_loss_threshold: float = 0.0, valid_loss_threshold: float = 0.0,
-              alpha: float = 1e-4, rho: float = 0.2,
-              lambda_l1: float = 1e-4, lambda_l2: float = 1e-4) -> LogisticRegression:
+              alpha: float = 1e-4, rho: float = 0.2) -> LogisticRegression:
         if balancing_dataset:
             min_length = min(len(positive_texts), len(negative_texts))
             positive_texts = positive_texts[:min_length]
@@ -52,10 +51,7 @@ class TextBinaryClassifierTrainer(BaseTrainer):
             total_loss = 0.0
             for batch_texts, batch_labels in train_loader:
                 outputs = self.model.forward(batch_texts)
-                loss = loss_function(outputs, batch_labels) + self.model.elastic_net(
-                    alpha=alpha, rho=rho,
-                    lambda_l1=lambda_l1, lambda_l2=lambda_l2
-                )
+                loss = loss_function(outputs, batch_labels) + self.model.elastic_net(alpha=alpha, rho=rho)
                 optimizer.zero_grad()
                 loss.backward()
                 optimizer.step()
@@ -66,10 +62,7 @@ class TextBinaryClassifierTrainer(BaseTrainer):
                     with torch.no_grad():
                         self.model.eval()
                         outputs = self.model.forward(batch_texts)
-                        loss = loss_function(outputs, batch_labels) + self.model.elastic_net(
-                            alpha=alpha, rho=rho,
-                            lambda_l1=lambda_l1, lambda_l2=lambda_l2
-                        )
+                        loss = loss_function(outputs, batch_labels) + self.model.elastic_net(alpha=alpha, rho=rho)
                         total_valid_loss += loss.item()
                         self.model.train()
                 logger.debug(f"Epoch {epoch + 1}/{num_epochs} | "

{deeplotx-0.2.18.dist-info → deeplotx-0.2.20.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deeplotx
-Version: 0.2.18
+Version: 0.2.20
 Summary: Easy-2-use long text classifier trainers.
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown

{deeplotx-0.2.18.dist-info → deeplotx-0.2.20.dist-info}/RECORD

@@ -3,18 +3,18 @@ deeplotx/encoder/__init__.py,sha256=x7k8IE0FXvDl7kCJGWPsetOHFdvNCiCXHbYOdvo7_JQ,
 deeplotx/encoder/bert_encoder.py,sha256=rdT8YgZzvRoqYqtzPW95ilagSQTAQgUl7mMVetGKxCY,1822
 deeplotx/encoder/long_text_encoder.py,sha256=yEEtTVZYHJ0W3OSbh7BHm6xI33nJmVYlSrgD5RVcJLY,2967
 deeplotx/nn/__init__.py,sha256=9gh8rhKqVWtJyvryU_wHPTLEQIorwOBhAQRc0DtNamM,153
-deeplotx/nn/base_neural_network.py,sha256=EjBqW_8DMNX9TqNzOaLsLi5RzgsCyBaCgwuVque4jwk,1580
+deeplotx/nn/base_neural_network.py,sha256=MXuID5bagdHyrFOkoybW1oiXAY2d4FGnzZoR37LZfUI,1566
 deeplotx/nn/linear_regression.py,sha256=D4mEWVOq6q1Fm2otm57rgZ_E06HJLZBV5k636PprAf4,1520
 deeplotx/nn/logistic_regression.py,sha256=QAtZp2oyqOW8-1pJWVcahsSM83bzfA68EHObg-wSHHY,463
 deeplotx/nn/softmax_regression.py,sha256=eUn3mVNlye9ewVdw3McPHZuKbUvvaamsUgFIJMVMgBU,487
 deeplotx/trainer/__init__.py,sha256=Fl5DR9UecQc5VtBcczU9sx_HtPNoFohpuELOh-Jrsks,77
 deeplotx/trainer/base_trainer.py,sha256=z0MeAT-rRYmjeBXt0ckt7J1itYArR0Cx02wHesXUoZE,385
-deeplotx/trainer/text_binary_classification_trainer.py,sha256=SEisKhDhcOoV87N0xLLVwuZrDSadcHmdoFIji-RqjT4,4375
+deeplotx/trainer/text_binary_classification_trainer.py,sha256=5O-5dwVMCj5EDX9gjJwCA468OR4UozJ7V8b-JxeUB0s,4080
 deeplotx/util/__init__.py,sha256=JxqAK_WOOHcYVSTHBT1-WuBwWrPEVDTV3titeVWvNUM,74
 deeplotx/util/hash.py,sha256=wwsC6kOQvbpuvwKsNQOARd78_wePmW9i3oaUuXRUnpc,352
 deeplotx/util/read_file.py,sha256=ptzouvEQeeW8KU5BrWNJlXw-vFXVrpS9SkAUxsu6A8A,612
-deeplotx-0.2.18.dist-info/licenses/LICENSE,sha256=IwGE9guuL-ryRPEKi6wFPI_zOhg7zDZbTYuHbSt_SAk,35823
-deeplotx-0.2.18.dist-info/METADATA,sha256=fn9tKJrvY8_JnnnpMG4N6EajaTI3HgN2iRrFodc5SE0,1573
-deeplotx-0.2.18.dist-info/WHEEL,sha256=0CuiUZ_p9E4cD6NyLD6UG80LBXYyiSYZOKDm5lp32xk,91
-deeplotx-0.2.18.dist-info/top_level.txt,sha256=hKg4pVDXZ-WWxkRfJFczRIll1Sv7VyfKCmzHLXbuh1U,9
-deeplotx-0.2.18.dist-info/RECORD,,
+deeplotx-0.2.20.dist-info/licenses/LICENSE,sha256=IwGE9guuL-ryRPEKi6wFPI_zOhg7zDZbTYuHbSt_SAk,35823
+deeplotx-0.2.20.dist-info/METADATA,sha256=NQgRWucDSAI4awAJNf9984IujFRo9PurR1qrqpmWIzA,1573
+deeplotx-0.2.20.dist-info/WHEEL,sha256=0CuiUZ_p9E4cD6NyLD6UG80LBXYyiSYZOKDm5lp32xk,91
+deeplotx-0.2.20.dist-info/top_level.txt,sha256=hKg4pVDXZ-WWxkRfJFczRIll1Sv7VyfKCmzHLXbuh1U,9
+deeplotx-0.2.20.dist-info/RECORD,,