radnn 0.0.8__py3-none-any.whl → 0.1.0__py3-none-any.whl

radnn/data/dataset_base_legacy.py

@@ -0,0 +1,280 @@
+# ......................................................................................
+# MIT License
+
+# Copyright (c) 2023 Pantelis I. Kaplanoglou
+
+# Permission is hereby granted, free of charge, to any person obtaining a copy
+# of this software and associated documentation files (the "Software"), to deal
+# in the Software without restriction, including without limitation the rights
+# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+# copies of the Software, and to permit persons to whom the Software is
+# furnished to do so, subject to the following conditions:
+
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+# SOFTWARE.
+
+# ......................................................................................
+
+
+import numpy as np
+from sklearn.model_selection import train_test_split  # import a standalone procedure function from the pacakge
+
+
+# =========================================================================================================================
+class CDataSetBase(object):
+  # --------------------------------------------------------------------------------------
+  # Constructor
+  def __init__(self, name=None, random_seed=None, is_classification=True):
+    # ................................................................
+    # // Fields \\
+    self.Name = name
+    self.RandomSeed = random_seed
+    self.IsClassification = is_classification
+    
+    self.FeatureCount = None
+    self.ClassCount = None
+    self.class_names = dict()
+    self.SampleCount = None
+    
+    self.Samples = None
+    self.Targets = None
+    
+    self.TSSampleIDs = None
+    self.TSSamples = None
+    self.TSTargets = None
+    self.TSSampleCount = 0
+    
+    self.VSSampleIDs = None
+    self.VSSamples = None
+    self.VSTargets = None
+    self.VSSampleCount = 0
+    
+    self.UTSampleIDs = None
+    self.UTSamples = None
+    self.UTTargets = None
+    self.UTSampleCount = None
+    
+    # ................................................................
+    if self.RandomSeed is not None:
+      ml.RandomSeed(self.RandomSeed)
+  
+  # --------------------------------------------------------------------------------------
+  def preview_images(self):
+    import matplotlib.pyplot as plt
+    
+    # Look at some sample images from dataset
+    plt.figure(figsize=(10, 10))
+    for i in range(25):
+      plt.subplot(5, 5, i + 1)
+      if self.TSSamples.shape[2] == 2:
+        plt.imshow(self.TSSamples[i].squeeze().astype(np.uint8), cmap='gray')
+      else:
+        plt.imshow(self.TSSamples[i].squeeze().astype(np.uint8))
+      
+      nClassIndex = self.TSLabels[i]
+      sClassDescr = str(nClassIndex)
+      if isinstance(self.class_names, dict):
+        if nClassIndex in self.class_names:
+          sClassName = self.class_names[nClassIndex]
+          sClassDescr += " `" + sClassName + "`"
+      elif isinstance(self.class_names, list):
+        sClassName = self.class_names[nClassIndex]
+        sClassDescr += " `" + sClassName + "` "
+      
+      plt.title(f"Label: {sClassDescr}")
+      plt.axis('off')
+    plt.show()
+  
+  # --------------------------------------------------------------------------------------
+  @property
+  def Labels(self):
+    return self.Targets
+  
+  # --------------------------------------------------------------------------------------
+  @property
+  def TSLabels(self):
+    return self.TSTargets
+  
+  # --------------------------------------------------------------------------------------
+  @property
+  def VSLabels(self):
+    return self.VSTargets
+  
+  # --------------------------------------------------------------------------------------
+  @property
+  def UTLabels(self):
+    return self.UTTargets
+  
+  # --------------------------------------------------------------------------------------
+  def TrainingSet(self, p_nSamples, p_nTargets):
+    self.TSSamples = p_nSamples
+    self.TSTargets = p_nTargets
+    self.countSamples()
+    self.TSSampleIDs = np.arange(0, self.TSSampleCount)
+    
+    # Feature count is calculated on samples that are flattened as vectors
+    if self.FeatureCount is None:
+      self.FeatureCount = np.prod(self.TSSamples.shape[1:])
+    
+    if self.ClassCount is None:
+      if self.IsClassification:
+        self.ClassCount = len(np.unique(self.TSTargets))
+      else:
+        self.ClassCount = 0
+  
+  # --------------------------------------------------------------------------------------
+  def ValidationSet(self, p_nSamples, p_nTargets):
+    self.VSSamples = p_nSamples
+    self.VSTargets = p_nTargets
+    self.countSamples()
+    self.VSSampleIDs = np.arange(0, self.VSSampleCount)
+  
+  # --------------------------------------------------------------------------------------
+  def UnknownTestSet(self, p_nSamples, p_nTargets):
+    self.UTSamples = p_nSamples
+    self.UTTargets = p_nTargets
+    self.countSamples()
+    self.UTSampleIDs = np.arange(0, self.UTSampleCount)
+    # --------------------------------------------------------------------------------------
+  
+  def info(self):
+    self.PrintInfo()
+  
+  # --------------------------------------------------------------------------------------
+  def PrintInfo(self):
+    print("Dataset [%s]" % self.Name)
+    print("  |__ FeatureCount:", self.FeatureCount)
+    if self.IsClassification:
+      print("  |__ ClassCount:", self.ClassCount)
+    
+    if self.TSSamples is not None:
+      print("  |__ Training set samples  : %d   shape:%s" % (self.TSSampleCount, self.TSSamples.shape))
+    if self.TSTargets is not None:
+      print("  |__ Training set targets  : %d   shape:%s" % (self.TSSampleCount, self.TSTargets.shape))
+    
+    if self.VSSamples is not None:
+      print("  |__ Validation set samples: %d   shape:%s" % (self.VSSampleCount, self.VSSamples.shape))
+    if self.VSTargets is not None:
+      print("  |__ Validation set targets: %d   shape:%s" % (self.VSSampleCount, self.VSTargets.shape))
+    
+    if self.UTSamples is not None:
+      print("  |__ MemoryTest set samples      : %d   shape:%s" % (self.UTSampleCount, self.UTSamples.shape))
+    if self.UTTargets is not None:
+      print("  |__ MemoryTest set targets      : %d   shape:%s" % (self.UTSampleCount, self.UTTargets.shape))
+  
+  # --------------------------------------------------------------------------------------
+  def countSamples(self):
+    if self.TSSamples is not None:
+      self.TSSampleCount = self.TSSamples.shape[0]
+      self.SampleCount = self.TSSampleCount + self.VSSampleCount
+    
+    if self.VSSamples is not None:
+      self.VSSampleCount = self.VSSamples.shape[0]
+      self.SampleCount = self.TSSampleCount + self.VSSampleCount
+      
+      # The test set samples are not included in the available sample count
+    if self.UTSamples is not None:
+      self.UTSampleCount = self.UTSamples.shape[0]
+  
+  # --------------------------------------------------------------------------------------
+  def Split(self, p_nTrainingPercentage):
+    nTSSamples, nVSSamples, nTSTargets, nVSTargets = train_test_split(self.Samples, self.Targets
+                                                                      , test_size=1.0 - p_nTrainingPercentage
+                                                                      , random_state=self.RandomSeed
+                                                                      , shuffle=True
+                                                                      , stratify=self.Targets
+                                                                      )
+    self.TrainingSet(nTSSamples, nTSTargets)
+    self.ValidationSet(nVSSamples, nVSTargets)
+    self.countSamples()
+  
+  # --------------------------------------------------------------------------------------
+  def LoadCache(self, p_oFileStore, p_sSamplesFilePrefix="Samples", p_sTargetsFilePrefix="Targets", p_bIsVerbose=False):
+    bResult = p_oFileStore.Exists("%s.pkl" % p_sSamplesFilePrefix) or p_oFileStore.Exists(
+      "%s.TS.pkl" % p_sSamplesFilePrefix)
+    
+    if bResult:
+      if p_bIsVerbose:
+        print("Loading known data set ...")
+      self.Samples = p_oFileStore.Deserialize("%s.pkl" % p_sSamplesFilePrefix)
+      self.Targets = p_oFileStore.Deserialize("%s.pkl" % p_sTargetsFilePrefix)
+      
+      if p_bIsVerbose:
+        print("Loading training set ...")
+      nTSSamples = p_oFileStore.Deserialize("%s.TS.pkl" % p_sSamplesFilePrefix)
+      nTSTargets = p_oFileStore.Deserialize("%s.TS.pkl" % p_sTargetsFilePrefix)
+      self.TrainingSet(nTSSamples, nTSTargets)
+      
+      if p_bIsVerbose:
+        print("Loading validation set ...")
+      nVSSamples = p_oFileStore.Deserialize("%s.VS.pkl" % p_sSamplesFilePrefix)
+      nVSTargets = p_oFileStore.Deserialize("%s.VS.pkl" % p_sTargetsFilePrefix)
+      self.ValidationSet(nVSSamples, nVSTargets)
+      
+      if p_bIsVerbose:
+        print("Loading unknown test data set ...")
+      nUTSamples = p_oFileStore.Deserialize("%s.UT.pkl" % p_sSamplesFilePrefix)
+      if nUTSamples is not None:
+        nUTTargets = p_oFileStore.Deserialize("%s.UT.pkl" % p_sTargetsFilePrefix)
+        self.UnknownTestSet(nUTSamples, nUTTargets)
+    
+    return bResult
+  
+  # --------------------------------------------------------------------------------------
+  def SaveCache(self, p_oFileStore, p_sSamplesFilePrefix="Samples", p_sTargetsFilePrefix="Targets"):
+    if self.Samples is not None:
+      p_oFileStore.Serialize("%s.pkl" % p_sSamplesFilePrefix, self.Samples, True)
+      p_oFileStore.Serialize("%s.pkl" % p_sTargetsFilePrefix, self.Labels, True)
+    
+    p_oFileStore.Serialize("%s.TS.pkl" % p_sSamplesFilePrefix, self.TSSamples, True)
+    p_oFileStore.Serialize("%s.TS.pkl" % p_sTargetsFilePrefix, self.TSLabels, True)
+    
+    p_oFileStore.Serialize("%s.VS.pkl" % p_sSamplesFilePrefix, self.VSSamples, True)
+    p_oFileStore.Serialize("%s.VS.pkl" % p_sTargetsFilePrefix, self.VSLabels, True)
+    
+    if self.UTSamples is not None:
+      p_oFileStore.Serialize("%s.UT.pkl" % p_sSamplesFilePrefix, self.UTSamples, True)
+      p_oFileStore.Serialize("%s.UT.pkl" % p_sTargetsFilePrefix, self.UTLabels, True)
+  # --------------------------------------------------------------------------------------
+
+
+# =========================================================================================================================
+
+
+if __name__ == "__main__":
+  ml.RandomSeed(2023)
+  
+  #nData = np.random.rand(100).astype(np.float32)
+  nData = np.arange(0, 100)
+  nData = nData / (nData + 1)
+  print(nData.shape)
+  nTargets = np.concatenate([np.zeros(50), np.ones(50)], axis=0).astype(np.int32)
+  print(nTargets.shape)
+  
+  oDataset = CDataSetBase("dummy")
+  oDataset.Samples = nData
+  oDataset.Targets = nTargets
+  oDataset.Split(0.8)
+  oDataset.PrintInfo()
+  
+  nTSChecksum = np.sum(oDataset.TSSamples)
+  nVSChecksum = np.sum(oDataset.VSSamples)
+  print(nTSChecksum, nVSChecksum)
+  
+  # // Unit Testing \\
+  assert nTSChecksum == 75.59110064555725
+  assert nVSChecksum == 19.22152183680314
+  
+  # Seed None:                 75.45033306926038 19.362289413100005
+  # Seed None: Parameter 2023: 75.59110064555725 19.22152183680314
+  # Seed 2023: Parameter 2023: 75.59110064555725 19.22152183680314  # First call with the same seed
+  # Seed 2023: Parameter None: 75.59110064555725 19.22152183680314  # First call with the same seed
+  # Seed 2025:                 75.55118094349115 19.261441538869224

radnn/data/errors.py

@@ -0,0 +1,32 @@
+# ......................................................................................
+# MIT License
+
+# Copyright (c) 2019-2025 Pantelis I. Kaplanoglou
+
+# Permission is hereby granted, free of charge, to any person obtaining a copy
+# of this software and associated documentation files (the "Software"), to deal
+# in the Software without restriction, including without limitation the rights
+# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+# copies of the Software, and to permit persons to whom the Software is
+# furnished to do so, subject to the following conditions:
+
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+# SOFTWARE.
+
+# ......................................................................................
+ERR_NO_CALLBACKS = "You should assign callbacks to the dataset perform proper random seed initialization for your framework."
+ERR_NO_RANDOM_SEED_INITIALIZER_CALLBACK = "Callback method for random seed initialization has not been defined."
+
+ERR_SUBSET_MUST_HAVE_TS = "A dataset must have at least a training subset."
+ERR_SUBSET_INVALID_SETUP = "Invalid sample subset setup. Please use one of the valid kinds: 'training/train/ts', 'validation/val/vs', 'testing/test/us'."
+ERR_SUBSET_MUST_HAVE_SAMPLES = "The subset has no samples, check the implementation of your dataset class."
+ERR_DATASET_FOLDER_NOT_FOUND = "The dataset was not found under the folder %s"
+ERR_DATASET_MUST_PROVIDE_LOCAL_FILESTORE = "You must provide a local filestore/path for the dataset"

radnn/data/sample_preprocessor.py

@@ -0,0 +1,58 @@
+# ......................................................................................
+# MIT License
+
+# Copyright (c) 2019-2025 Pantelis I. Kaplanoglou
+
+# Permission is hereby granted, free of charge, to any person obtaining a copy
+# of this software and associated documentation files (the "Software"), to deal
+# in the Software without restriction, including without limitation the rights
+# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+# copies of the Software, and to permit persons to whom the Software is
+# furnished to do so, subject to the following conditions:
+
+# The above copyright notice and this permission notice shall be included in all
+# copies or substantial portions of the Software.
+
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+# SOFTWARE.
+
+# ......................................................................................
+from abc import ABC, abstractmethod
+
+
+# ======================================================================================================================
+class SamplePreprocessor(ABC):
+  # --------------------------------------------------------------------------------------------------------------------
+  def __init__(self, settings):
+    self.settings = settings
+    self.ts_transform_augment, self.vs_transform_augment, self.us_transform_augment = self.prepare_pipelines()
+  # --------------------------------------------------------------------------------------------------------------------
+  def __call__(self, samples, **kwargs):
+    if self.us_transform_augment is None:
+      return samples
+    else:
+      return self.input_for_inference(input)
+  # --------------------------------------------------------------------------------------------------------------------
+  @abstractmethod
+  def input_for_inference(self, samples):
+    pass
+  # --------------------------------------------------------------------------------------------------------------------
+  @abstractmethod
+  def prepare_pipelines(self):
+    pass
+  # --------------------------------------------------------------------------------------------------------------------
+
+# ======================================================================================================================
+class VoidPreprocessor(SamplePreprocessor):
+  # --------------------------------------------------------------------------------------------------------------------
+  def input_for_inference(self, samples):
+    return samples
+  # --------------------------------------------------------------------------------------------------------------------
+  def prepare_pipelines(self):
+    return None, None, None
+  # --------------------------------------------------------------------------------------------------------------------