[TYPING](ALL): Fix typing errors

biobb_pytorch/docs/source/conf.py

@@ -13,7 +13,6 @@
 # serve to show the default.
 
 import sys
-from typing import List, Dict
 from pathlib import Path
 
 # If extensions (or modules to document with autodoc) are in another directory,
@@ -95,9 +94,9 @@
 # Else, today_fmt is used as the format for a strftime call.
 # today_fmt = '%B %d, %Y'
 
-# List of patterns, relative to source directory, that match files and
+# list of patterns, relative to source directory, that match files and
 # directories to ignore when looking for source files.
-exclude_patterns: List[str] = []
+exclude_patterns: list[str] = []
 
 # The reST default role (used for this markup: `text`) to use for all
 # documents.
@@ -123,7 +122,7 @@
 # If true, keep warnings as "system message" paragraphs in the built documents.
 # keep_warnings = False
 
-# If true, `todo` and `todoList` produce output, else they produce nothing.
+# If true, `todo` and `todolist` produce output, else they produce nothing.
 todo_include_todos = False
 
 
@@ -232,7 +231,7 @@ def setup(app):
 
 # -- Options for LaTeX output ---------------------------------------------
 
-latex_elements: Dict[str, str] = {
+latex_elements: dict[str, str] = {
     # The paper size ('letterpaper' or 'a4paper').
     # 'papersize': 'letterpaper',
 
@@ -246,7 +245,7 @@ def setup(app):
     # 'figure_align': 'htbp',
 }
 
-# Grouping the document tree into LaTeX files. List of tuples
+# Grouping the document tree into LaTeX files. list of tuples
 # (source start file, target name, title,
 #  author, documentclass [howto, manual, or own class]).
 latex_documents = [
@@ -277,7 +276,7 @@ def setup(app):
 
 # -- Options for manual page output ---------------------------------------
 
-# One entry per manual page. List of tuples
+# One entry per manual page. list of tuples
 # (source start file, name, description, authors, manual section).
 man_pages = [
     (master_doc, 'biobb_pytorch', u'biobb_pytorch Documentation',
@@ -290,7 +289,7 @@ def setup(app):
 
 # -- Options for Texinfo output -------------------------------------------
 
-# Grouping the document tree into Texinfo files. List of tuples
+# Grouping the document tree into Texinfo files. list of tuples
 # (source start file, target name, title, author,
 #  dir menu entry, description, category)
 texinfo_documents = [

biobb_pytorch/mdae/apply_mdae.py

@@ -2,10 +2,12 @@
 
 """Module containing the ApplyMDAE class and the command line interface."""
 import torch
+import torch.utils.data
 import numpy as np
 import argparse
+from typing import Optional
 import time
-from typing import Optional, Tuple, Dict
+from typing import Optional
 from biobb_common.generic.biobb_object import BiobbObject
 from biobb_common.configuration import settings
 from biobb_common.tools import file_utils as fu
@@ -60,7 +62,7 @@ def __init__(self,
                  input_model_pth_path: str,
                  output_reconstructed_data_npy_path: str,
                  output_latent_space_npy_path: Optional[str] = None,
-                 properties: Optional[Dict] = None, **kwargs) -> None:
+                 properties: Optional[dict] = None, **kwargs) -> None:
         properties = properties or {}
 
         # Call parent class constructor
@@ -131,7 +133,7 @@ def launch(self) -> int:
         self.check_arguments(output_files_created=True, raise_exception=False)
         return 0
 
-    def apply_model(self, dataloader: torch.utils.data.DataLoader) -> Tuple[np.ndarray, np.ndarray]:
+    def apply_model(self, dataloader: torch.utils.data.DataLoader) -> tuple[np.ndarray, np.ndarray]:
         self.model.to(self.model.device)
         start_time: float = time.time()
         fu.log("Applying model:", self.out_log)
@@ -153,7 +155,7 @@ def apply_model(self, dataloader: torch.utils.data.DataLoader) -> Tuple[np.ndarr
 
 def applyMDAE(input_data_npy_path: str, input_model_pth_path: str,
               output_reconstructed_data_npy_path: str, output_latent_space_npy_path: Optional[str] = None,
-              properties: Optional[Dict] = None, **kwargs) -> int:
+              properties: Optional[dict] = None, **kwargs) -> int:
     """Execute the :class:`ApplyMDAE <mdae.apply_mdae.ApplyMDAE>` class and
     execute the :meth:`launch() <mdae.apply_mdae.ApplyMDAE.launch>` method."""
 

biobb_pytorch/mdae/common.py

@@ -2,7 +2,7 @@
 import numpy as np
 import torch
 from pathlib import Path
-from typing import Callable, List, Optional, Tuple, Union
+from typing import Callable, Optional, Union
 
 
 def ndarray_normalization(ndarray: np.ndarray, max_values: np.ndarray, min_values: np.ndarray) -> np.ndarray:
@@ -69,11 +69,11 @@ def get_optimizer_function(optimizer_function: str) -> Callable:
         raise ValueError(f'Invalid optimizer function: {optimizer_function}')
 
 
-def execute_model(model: torch.nn.Module, dataloader: torch.utils.data.DataLoader, input_dimensions: int, latent_dimensions: int, loss_function: Optional[torch.nn.modules.loss._Loss] = None) -> Tuple[float, np.ndarray, np.ndarray]:
+def execute_model(model: torch.nn.Module, dataloader: torch.utils.data.DataLoader, input_dimensions: int, latent_dimensions: int, loss_function: Optional[torch.nn.modules.loss._Loss] = None) -> tuple[float, np.ndarray, np.ndarray]:
     model.eval()
-    losses: List[float] = []
-    z_list: List[float] = []
-    x_hat_list: List[float] = []
+    losses: list[float] = []
+    z_list: list[float] = []
+    x_hat_list: list[float] = []
     with torch.no_grad():
         for data in dataloader:
             data = data[0].to(model.device)

biobb_pytorch/mdae/mdae.py

@@ -1,6 +1,5 @@
 """Module containing the MDAutoEncoder class and the command line interface."""
 import torch
-from typing import List
 
 
 class MDAE(torch.nn.Module):
@@ -16,7 +15,7 @@ def __init__(self, input_dimensions: int, num_layers: int, latent_dimensions: in
         self.leaky_relu: float = leaky_relu
 
         # Encoder
-        encoder: List = []
+        encoder: list = []
         nunits: int = self.input_dimensions
         for _ in range(self.num_layers-1):
             encoder.append(torch.nn.Linear(nunits, nunits - self.delta))
@@ -32,7 +31,7 @@ def __init__(self, input_dimensions: int, num_layers: int, latent_dimensions: in
             torch.nn.Sigmoid())
 
         # Decoder
-        decoder: List = []
+        decoder: list = []
         nunits = self.latent_dimensions
         for _ in range(self.num_layers-1):
             decoder.append(torch.nn.Linear(nunits, nunits + self.delta))

biobb_pytorch/mdae/train_mdae.py

@@ -2,10 +2,14 @@
 
 """Module containing the TrainMDAE class and the command line interface."""
 import torch
+import torch.utils.data
 import numpy as np
 import time
 import argparse
-from typing import Optional, List, Tuple, Dict
+from typing import Optional
+from typing import Optional
+from torch.optim.optimizer import Optimizer
+from torch.optim.adam import Adam
 from biobb_common.generic.biobb_object import BiobbObject
 from biobb_common.configuration import settings
 from biobb_common.tools import file_utils as fu
@@ -87,7 +91,7 @@ def __init__(self, input_train_npy_path: str,
                  input_model_pth_path: Optional[str] = None,
                  output_train_data_npz_path: Optional[str] = None,  # npz of  train_losses, valid_losses
                  output_performance_npz_path: Optional[str] = None,  # npz of  evaluate_losses, latent_space, reconstructed_data
-                 properties: Optional[Dict] = None, **kwargs) -> None:
+                 properties: Optional[dict] = None, **kwargs) -> None:
         properties = properties or {}
 
         # Call parent class constructor
@@ -177,12 +181,11 @@ def __init__(self, input_train_npy_path: str,
 
         optimizer_str: str = properties.get('optimizer', '')
         try:
-            self.optimizer: torch.optim.Optimizer = get_optimizer_function(optimizer_str)(self.model.parameters(), lr=self.lr)
+            self.optimizer = get_optimizer_function(optimizer_str)(self.model.parameters(), lr=self.lr)
             fu.log(f'Using optimizer: {self.optimizer}', self.out_log)
         except ValueError:
             fu.log(f'Invalid optimizer: {optimizer_str}', self.out_log)
-            fu.log('Using default optimizer: Adam', self.out_log)
-            self.optimizer = torch.optim.Adam(self.model.parameters(), lr=self.lr)
+            self.optimizer = Adam(self.model.parameters(), lr=self.lr)
 
     @launchlogger
     def launch(self) -> int:
@@ -225,10 +228,10 @@ def launch(self) -> int:
         self.check_arguments(output_files_created=True, raise_exception=False)
         return 0
 
-    def train_model(self) -> Tuple[List[float], List[float], Dict, int]:
+    def train_model(self) -> tuple[list[float], list[float], dict, int]:
         self.model.to(self.model.device)
-        train_losses: List[float] = []
-        validation_losses: List[float] = []
+        train_losses: list[float] = []
+        validation_losses: list[float] = []
         best_valid_loss: float = float('inf')  # Initialize best valid loss to infinity
 
         start_time: float = time.time()
@@ -280,16 +283,16 @@ def train_model(self) -> Tuple[List[float], List[float], Dict, int]:
             # Save best model
             if avg_validation_loss < best_valid_loss:
                 best_valid_loss = avg_validation_loss
-                best_model: Dict = self.model.state_dict()
+                best_model: dict = self.model.state_dict()
                 best_model_epoch: int = epoch_index
 
         fu.log(f"End Training, total time: {format_time((time.time() - start_time))}", self.out_log)
 
         return train_losses, validation_losses, best_model, best_model_epoch
 
-    def training_step(self, dataloader: torch.utils.data.DataLoader, optimizer: torch.optim.Optimizer, loss_function: torch.nn.modules.loss._Loss) -> Tuple[float, float]:
+    def training_step(self, dataloader: torch.utils.data.DataLoader, optimizer: Optimizer, loss_function: torch.nn.modules.loss._Loss) -> tuple[float, float]:
         self.model.train()
-        train_losses: List[float] = []
+        train_losses: list[float] = []
         for data in dataloader:
             data = data[0].to(self.model.device)
             _, output = self.model(data)
@@ -300,7 +303,7 @@ def training_step(self, dataloader: torch.utils.data.DataLoader, optimizer: torc
             train_losses.append(loss.item())
 
         self.model.eval()
-        valid_losses: List[float] = []
+        valid_losses: list[float] = []
         with torch.no_grad():
             for data in dataloader:
                 data = data[0].to(self.model.device)
@@ -310,13 +313,13 @@ def training_step(self, dataloader: torch.utils.data.DataLoader, optimizer: torc
 
         return float(np.mean(train_losses)), float(torch.mean(torch.tensor(valid_losses)))
 
-    def evaluate_model(self, dataloader: torch.utils.data.DataLoader, loss_function: torch.nn.modules.loss._Loss) -> Tuple[float, np.ndarray, np.ndarray]:
+    def evaluate_model(self, dataloader: torch.utils.data.DataLoader, loss_function: torch.nn.modules.loss._Loss) -> tuple[float, np.ndarray, np.ndarray]:
         return execute_model(self.model, dataloader, self.input_dimensions, self.latent_dimensions, loss_function)
 
 
 def trainMDAE(input_train_npy_path: str, output_model_pth_path: str, input_model_pth_path: Optional[str] = None,
               output_train_data_npz_path: Optional[str] = None, output_performance_npz_path: Optional[str] = None,
-              properties: Optional[Dict] = None, **kwargs) -> int:
+              properties: Optional[dict] = None, **kwargs) -> int:
     """Execute the :class:`TrainMDAE <mdae.train_mdae.TrainMDAE>` class and
     execute the :meth:`launch() <mdae.train_mdae.TrainMDAE.launch>` method."""