ddi-fw 0.0.212__tar.gz → 0.0.214__tar.gz

{ddi_fw-0.0.212 → ddi_fw-0.0.214}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ddi_fw
-Version: 0.0.212
+Version: 0.0.214
 Summary: Do not use :)
 Author-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>
 Maintainer-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>

{ddi_fw-0.0.212 → ddi_fw-0.0.214}/pyproject.toml

@@ -6,7 +6,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "ddi_fw"
-version = "0.0.212"
+version = "0.0.214"
 description = "Do not use :)"
 readme = "README.md"
 authors = [

{ddi_fw-0.0.212 → ddi_fw-0.0.214}/src/ddi_fw/ml/ml_helper.py

@@ -156,7 +156,7 @@ class MultiModalRunner:
             (self.y_test_label.shape[0], self.y_test_label.shape[1]))
         for item in combination:
             prediction = prediction + single_results[item]
-        prediction = np.argmax(prediction, axis=1)
+        prediction = utils.to_one_hot_encode(prediction)
         logs, metrics = evaluate(
             actual=self.y_test_label, pred=prediction, info=combination_descriptor)
         if self.use_mlflow:

{ddi_fw-0.0.212 → ddi_fw-0.0.214}/src/ddi_fw/ml/tensorflow_wrapper.py

@@ -183,11 +183,11 @@ class TFModelWrapper(ModelWrapper):
                 actual = self.test_label
                 # if not utils.is_binary_encoded(pred):
                 #     pred = tf.keras.utils.to_categorical(np.argmax(pred,axis=1), num_classes=self.num_classes)
-                pred= convert_to_categorical(pred, self.num_classes)
-                actual= convert_to_categorical(actual, self.num_classes)
+                pred_as_cat= convert_to_categorical(pred, self.num_classes)
+                actual_as_cat= convert_to_categorical(actual, self.num_classes)
 
                 logs, metrics = evaluate(
-                    actual=actual, pred=pred, info=self.descriptor, print_detail=print_detail)
+                    actual=actual_as_cat, pred=pred_as_cat, info=self.descriptor, print_detail=print_detail)
                 metrics.format_float()
                 mlflow.log_metrics(logs)
                 mlflow.log_param('best_cv', best_model_key)

{ddi_fw-0.0.212 → ddi_fw-0.0.214}/src/ddi_fw/utils/__init__.py

@@ -4,5 +4,5 @@ from .py7zr_helper import Py7ZipHelper
 from .enums import UMLSCodeTypes, DrugBankTextDataTypes
 from .package_helper import get_import
 from .kaggle import create_kaggle_dataset
-from .categorical_data_encoding_checker import is_one_hot_encoded, is_binary_encoded, is_binary_vector,is_label_encoded
+from .categorical_data_encoding_checker import to_one_hot_encode,is_one_hot_encoded, is_binary_encoded, is_binary_vector,is_label_encoded
 from .numpy_utils import adjust_array_dims

ddi_fw-0.0.214/src/ddi_fw/utils/categorical_data_encoding_checker.py

@@ -0,0 +1,90 @@
+import numpy as np
+
+def to_one_hot_encode(arr):
+    """
+    Convert a multi-dimensional array (1D, 2D, or 3D) into a one-hot encoded array.
+    
+    Parameters:
+    arr (numpy.ndarray): An array where each element or row (or slice) contains class indices or class probabilities.
+    
+    Returns:
+    numpy.ndarray: One-hot encoded array.
+    """
+    # Check if the input is a numpy array
+    if not isinstance(arr, np.ndarray):
+        raise ValueError("Input must be a numpy array")
+
+    # Get the shape of the input
+    shape = arr.shape
+    
+    # If the array is 1D, treat it as a list of class indices
+    if arr.ndim == 1:
+        num_classes = np.max(arr) + 1
+        one_hot = np.zeros((shape[0], num_classes))
+        one_hot[np.arange(shape[0]), arr] = 1
+
+    # If the array is 2D, treat each row as a list of class indices
+    elif arr.ndim == 2:
+        num_classes = shape[1]
+        one_hot = np.zeros((shape[0], num_classes))
+        # Handle one-hot encoding for each row (max index for each row)
+        max_indices = np.argmax(arr, axis=1)
+        one_hot[np.arange(shape[0]), max_indices] = 1
+
+    # If the array is 3D or higher, iterate over the first axis and apply one-hot encoding for each slice
+    elif arr.ndim >= 3:
+        num_classes = shape[-1]
+        # Initialize the output array for one-hot encoding with the same shape as input, but the last dimension is num_classes
+        one_hot = np.zeros_like(arr, dtype=int)
+        
+        # Iterate through the first axis (the batch dimension)
+        for i in range(shape[0]):
+            # Get the max indices along the second dimension for each slice
+            max_indices = np.argmax(arr[i], axis=1)
+            # Set the corresponding one-hot vectors in the slice
+            one_hot[i, np.arange(shape[1]), max_indices] = 1
+
+    return one_hot
+
+
+def is_one_hot_encoded(arr):
+    # Check if the array is one-hot encoded
+    # Ensure the input is a numpy ndarray and is 2D
+    if not isinstance(arr, np.ndarray):
+        raise ValueError("Input must be a NumPy ndarray.")
+    if not np.all(np.isin(arr, [0, 1])):
+        return False
+    # Check if each row (or column) has exactly one "1"
+    return np.all(np.sum(arr, axis=1) == 1)  # For row-wise checking
+
+
+def is_binary_encoded(arr):
+    # Ensure the input is a numpy ndarray and is 2D
+    if not isinstance(arr, np.ndarray):
+        raise ValueError("Input must be a NumPy ndarray.")
+    if arr.ndim != 2:
+        raise ValueError("Input must be a 2D array.")
+
+    # Check if all elements are either 0 or 1
+    return np.all(np.isin(arr, [0, 1]))
+
+
+def is_binary_vector(arr):
+    # Ensure the input is a numpy ndarray and is 1D
+    if not isinstance(arr, np.ndarray):
+        raise ValueError("Input must be a NumPy ndarray.")
+    if arr.ndim != 1:
+        raise ValueError("Input must be a 1D array.")
+    return arr.ndim == 1 and np.all(np.isin(arr, [0, 1]))
+
+
+def is_label_encoded(arr):
+    # Check if the array is label encoded
+    # Ensure the input is a numpy ndarray and is 1D
+    if not isinstance(arr, np.ndarray):
+        raise ValueError("Input must be a NumPy ndarray.")
+    if arr.ndim != 1:
+        raise ValueError("Input must be a 1D array.")
+    
+    # Check if all values are non-negative integers (possible class labels)
+    return np.issubdtype(arr.dtype, np.integer) and np.all(arr >= 0)

{ddi_fw-0.0.212 → ddi_fw-0.0.214}/src/ddi_fw.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ddi_fw
-Version: 0.0.212
+Version: 0.0.214
 Summary: Do not use :)
 Author-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>
 Maintainer-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>

ddi_fw-0.0.212/src/ddi_fw/utils/categorical_data_encoding_checker.py

@@ -1,44 +0,0 @@
-import numpy as np
-
-
-def is_one_hot_encoded(arr):
-    # Check if the array is one-hot encoded
-    # Ensure the input is a numpy ndarray and is 2D
-    if not isinstance(arr, np.ndarray):
-        raise ValueError("Input must be a NumPy ndarray.")
-    if not np.all(np.isin(arr, [0, 1])):
-        return False
-    # Check if each row (or column) has exactly one "1"
-    return np.all(np.sum(arr, axis=1) == 1)  # For row-wise checking
-
-
-def is_binary_encoded(arr):
-    # Ensure the input is a numpy ndarray and is 2D
-    if not isinstance(arr, np.ndarray):
-        raise ValueError("Input must be a NumPy ndarray.")
-    if arr.ndim != 2:
-        raise ValueError("Input must be a 2D array.")
-
-    # Check if all elements are either 0 or 1
-    return np.all(np.isin(arr, [0, 1]))
-
-
-def is_binary_vector(arr):
-    # Ensure the input is a numpy ndarray and is 1D
-    if not isinstance(arr, np.ndarray):
-        raise ValueError("Input must be a NumPy ndarray.")
-    if arr.ndim != 1:
-        raise ValueError("Input must be a 1D array.")
-    return arr.ndim == 1 and np.all(np.isin(arr, [0, 1]))
-
-
-def is_label_encoded(arr):
-    # Check if the array is label encoded
-    # Ensure the input is a numpy ndarray and is 1D
-    if not isinstance(arr, np.ndarray):
-        raise ValueError("Input must be a NumPy ndarray.")
-    if arr.ndim != 1:
-        raise ValueError("Input must be a 1D array.")
-    
-    # Check if all values are non-negative integers (possible class labels)
-    return np.issubdtype(arr.dtype, np.integer) and np.all(arr >= 0)