Fix CSAI cannot accept dataset files for lazy loading

pypots/classification/csai/core.py

@@ -11,28 +11,6 @@
 
 from ...nn.modules.csai import BackboneBCSAI
 
-# class DiceBCELoss(nn.Module):
-#     def __init__(self, weight=None, size_average=True):
-#         super(DiceBCELoss, self).__init__()
-#         self.bcelogits = nn.BCEWithLogitsLoss()
-
-#     def forward(self, y_score, y_out, targets, smooth=1):
-
-#         #comment out if your model contains a sigmoid or equivalent activation layer
-#         # inputs = F.sigmoid(inputs)
-
-#         #flatten label and prediction tensors
-#         BCE = self.bcelogits(y_out, targets)
-
-#         y_score = y_score.view(-1)
-#         targets = targets.view(-1)
-#         intersection = (y_score * targets).sum()
-#         dice_loss = 1 - (2.*intersection + smooth)/(y_score.sum() + targets.sum() + smooth)
-
-#         Dice_BCE = BCE + dice_loss
-
-#         return BCE, Dice_BCE
-
 
 class _BCSAI(nn.Module):
     def __init__(

pypots/classification/csai/data.py

@@ -6,6 +6,7 @@
 # License: BSD-3-Clause
 
 from typing import Union
+
 from ...imputation.csai.data import DatasetForCSAI as DatasetForCSAI_Imputation
 
 

pypots/classification/csai/model.py

@@ -6,15 +6,18 @@
 # License: BSD-3-Clause
 
 from typing import Optional, Union
-import numpy as np
+
 import torch
 from torch.utils.data import DataLoader
 
 from .core import _BCSAI
 from .data import DatasetForCSAI
 from ..base import BaseNNClassifier
+from ...data.checking import key_in_data_set
+from ...data.saving.h5 import load_dict_from_h5
 from ...optim.adam import Adam
 from ...optim.base import Optimizer
+from ...utils.logging import logger
 
 
 class CSAI(BaseNNClassifier):
@@ -60,31 +63,43 @@ class CSAI(BaseNNClassifier):
     The batch size for training and evaluating the model.
 
     epochs :
-    The number of epochs for training the model.
-
-    dropout :
-    The dropout rate for the model to prevent overfitting. Default is 0.5.
+        The number of epochs for training the model.
 
     patience :
-    The patience for the early-stopping mechanism. Given a positive integer, the training process will be stopped when the model does not perform better after that number of epochs. Leaving it default as None will disable the early-stopping.
+        The patience for the early-stopping mechanism. Given a positive integer, the training process will be
+        stopped when the model does not perform better after that number of epochs.
+        Leaving it default as None will disable the early-stopping.
 
     optimizer :
-    The optimizer for model training. If not given, will use a default Adam optimizer.
+        The optimizer for model training.
+        If not given, will use a default Adam optimizer.
 
     num_workers :
-    The number of subprocesses to use for data loading. 0 means data loading will be in the main process, i.e. there won't be subprocesses.
+        The number of subprocesses to use for data loading.
+        `0` means data loading will be in the main process, i.e. there won't be subprocesses.
 
     device :
-    The device for the model to run on. It can be a string, a :class:torch.device object, or a list of them. If not given, will try to use CUDA devices first (will use the default CUDA device if there are multiple), then CPUs, considering CUDA and CPU are so far the main devices for people to train ML models. If given a list of devices, e.g. ['cuda:0', 'cuda:1'], or [torch.device('cuda:0'), torch.device('cuda:1')], the model will be parallely trained on the multiple devices (so far only support parallel training on CUDA devices). Other devices like Google TPU and Apple Silicon accelerator MPS may be added in the future.
+        The device for the model to run on. It can be a string, a :class:`torch.device` object, or a list of them.
+        If not given, will try to use CUDA devices first (will use the default CUDA device if there are multiple),
+        then CPUs, considering CUDA and CPU are so far the main devices for people to train ML models.
+        If given a list of devices, e.g. ['cuda:0', 'cuda:1'], or [torch.device('cuda:0'), torch.device('cuda:1')] , the
+        model will be parallely trained on the multiple devices (so far only support parallel training on CUDA devices).
+        Other devices like Google TPU and Apple Silicon accelerator MPS may be added in the future.
 
     saving_path :
-    The path for automatically saving model checkpoints and tensorboard files (i.e. loss values recorded during training into a tensorboard file). Will not save if not given.
+        The path for automatically saving model checkpoints and tensorboard files (i.e. loss values recorded during
+        training into a tensorboard file). Will not save if not given.
 
     model_saving_strategy :
-    The strategy to save model checkpoints. It has to be one of [None, "best", "better", "all"]. No model will be saved when it is set as None. The "best" strategy will only automatically save the best model after the training finished. The "better" strategy will automatically save the model during training whenever the model performs better than in previous epochs. The "all" strategy will save every model after each epoch training.
+        The strategy to save model checkpoints. It has to be one of [None, "best", "better", "all"].
+        No model will be saved when it is set as None.
+        The "best" strategy will only automatically save the best model after the training finished.
+        The "better" strategy will automatically save the model during training whenever the model performs
+        better than in previous epochs.
+        The "all" strategy will save every model after each epoch training.
 
     verbose :
-    Whether to print out the training logs during the training process.
+        Whether to print out the training logs during the training process.
 
     """
 
@@ -101,10 +116,10 @@ def __init__(
         increase_factor: float,
         compute_intervals: bool,
         step_channels: int,
-        batch_size: int,
-        epochs: int,
         dropout: float = 0.5,
-        patience: Union[int, None] = None,
+        batch_size: int = 32,
+        epochs: int = 100,
+        patience: Optional[int] = None,
         optimizer: Optimizer = Adam(),
         num_workers: int = 0,
         device: Optional[Union[str, torch.device, list]] = None,
@@ -136,6 +151,9 @@ def __init__(
         self.compute_intervals = compute_intervals
         self.dropout = dropout
         self.intervals = None
+        self.replacement_probabilities = None
+        self.mean_set = None
+        self.std_set = None
 
         # Initialise empty model
         self.model = _BCSAI(
@@ -156,6 +174,7 @@ def __init__(
 
         # set up the optimizer
         self.optimizer = optimizer
+        self.optimizer.init_optimizer(self.model.parameters())
 
     def _assemble_input_for_training(self, data: list, training=True) -> dict:
         # extract data
@@ -230,7 +249,13 @@ def fit(
         file_type: str = "hdf5",
     ) -> None:
         # Create dataset
-        self.training_set = DatasetForCSAI(
+        if isinstance(train_set, str):
+            logger.warning(
+                "CSAI does not support lazy loading because normalise mean and std need to be calculated ahead. "
+                "Hence the whole train set will be loaded into memory."
+            )
+            train_set = load_dict_from_h5(train_set)
+        training_set = DatasetForCSAI(
             data=train_set,
             file_type=file_type,
             return_y=True,
@@ -239,19 +264,28 @@ def fit(
             compute_intervals=self.compute_intervals,
         )
 
-        self.intervals = self.training_set.intervals
-        self.replacement_probabilities = self.training_set.replacement_probabilities
-        self.mean_set = self.training_set.mean_set
-        self.std_set = self.training_set.std_set
+        self.intervals = training_set.intervals
+        self.replacement_probabilities = training_set.replacement_probabilities
+        self.mean_set = training_set.mean_set
+        self.std_set = training_set.std_set
 
         train_loader = DataLoader(
-            self.training_set,
+            training_set,
             batch_size=self.batch_size,
             shuffle=True,
             num_workers=self.num_workers,
         )
         val_loader = None
         if val_set is not None:
+            if isinstance(val_set, str):
+                logger.warning(
+                    "CSAI does not support lazy loading because normalise mean and std need to be calculated ahead. "
+                    "Hence the whole val set will be loaded into memory."
+                )
+                val_set = load_dict_from_h5(val_set)
+
+            if not key_in_data_set("X_ori", val_set):
+                raise ValueError("val_set must contain 'X_ori' for model validation.")
             val_set = DatasetForCSAI(
                 data=val_set,
                 file_type=file_type,
@@ -269,24 +303,6 @@ def fit(
                 shuffle=False,
                 num_workers=self.num_workers,
             )
-        # Create model
-        self.model = _BCSAI(
-            n_steps=self.n_steps,
-            n_features=self.n_features,
-            rnn_hidden_size=self.rnn_hidden_size,
-            imputation_weight=self.imputation_weight,
-            consistency_weight=self.consistency_weight,
-            classification_weight=self.classification_weight,
-            n_classes=self.n_classes,
-            step_channels=self.step_channels,
-            dropout=self.dropout,
-            intervals=self.intervals,
-        )
-        self._send_model_to_given_device()
-        self._print_model_size()
-
-        # set up the optimizer
-        self.optimizer.init_optimizer(self.model.parameters())
 
         # train the model
         self._train_model(train_loader, val_loader)
@@ -302,6 +318,13 @@ def predict(
     ) -> dict:
 
         self.model.eval()
+
+        if isinstance(test_set, str):
+            logger.warning(
+                "CSAI does not support lazy loading because normalise mean and std need to be calculated ahead. "
+                "Hence the whole test set will be loaded into memory."
+            )
+            test_set = load_dict_from_h5(test_set)
         test_set = DatasetForCSAI(
             data=test_set,
             file_type=file_type,
@@ -321,15 +344,15 @@ def predict(
             num_workers=self.num_workers,
         )
 
-        classificaion_results = []
+        classification_results = []
 
         with torch.no_grad():
             for idx, data in enumerate(test_loader):
                 inputs = self._assemble_input_for_testing(data)
                 results = self.model.forward(inputs, training=False)
-                classificaion_results.append(results["classification_pred"])
+                classification_results.append(results["classification_pred"])
 
-        classification = torch.cat(classificaion_results).cpu().detach().numpy()
+        classification = torch.cat(classification_results).cpu().detach().numpy()
         result_dict = {
             "classification": classification,
         }

pypots/imputation/csai/core.py

@@ -62,7 +62,10 @@ class _BCSAI(nn.Module):
 
     Notes
     -----
-    BCSAI is a bidirectional imputation model that uses forward and backward GRU cells to handle time-series data. It computes consistency and reconstruction losses to improve imputation accuracy. During training, the forward and backward reconstructions are combined, and losses are used to update the model. In evaluation mode, the model also outputs original data and indicating masks for further analysis.
+    CSAI is a bidirectional imputation model that uses forward and backward GRU cells to handle time-series data.
+    It computes consistency and reconstruction losses to improve imputation accuracy.
+    During training, the forward and backward reconstructions are combined, and losses are used to update the model.
+    In evaluation mode, the model also outputs original data and indicating masks for further analysis.
 
     """