pyerualjetwork 2.8.1__py3-none-any.whl → 2.8.2__py3-none-any.whl

plan/__init__.py

@@ -2,4 +2,4 @@
 
 # Bu dosya, plan modülünün ana giriş noktasıdır.
 
-from .plan import auto_balancer, normalization, Softmax, Sigmoid, Relu, weight_identification, fex, fit, evaluate, save_model, load_model, predict_model_ssd, predict_model_ram, get_weights, get_df, get_preds, get_acc, synthetic_augmentation, standard_scaler, multiple_evaluate, encode_one_hot, split, metrics, decode_one_hot, roc_curve, confusion_matrix, plot_evaluate, manuel_balancer, weight_normalization, plot_decision_space
+from .plan import auto_balancer, normalization, Softmax, Sigmoid, Relu, weight_identification, fex, fit, evaluate, save_model, load_model, predict_model_ssd, predict_model_ram, get_weights, get_df, get_preds, get_acc, synthetic_augmentation, standard_scaler, multiple_evaluate, encode_one_hot, split, metrics, decode_one_hot, roc_curve, confusion_matrix, plot_evaluate, manuel_balancer, weight_normalization, plot_decision_boundry

plan/plan.py

@@ -14,6 +14,7 @@ from scipy.special import expit, softmax
 import matplotlib.pyplot as plt
 import seaborn as sns
 from tqdm import tqdm
+from scipy.spatial import ConvexHull
 
 # BUILD -----
 
@@ -150,8 +151,9 @@ def fit(
             val_acc = validation_model[get_acc()]
             val_preds = validation_model[get_preds()]
             
+            plt.clf()
             
-            plot_decision_space(x_val, val_preds, s=100, color='tab20')
+            plot_decision_boundry(x_val, y_val, y_preds=val_preds, s=100, color='tab20')
 
 
             plt.pause(0.0001)
@@ -255,7 +257,7 @@ def fit(
         val_preds = validation_model[get_preds()]
         
         
-        plot_decision_space(x_val, val_preds, s=100, color='tab20')
+        plot_decision_boundry(x_val, y_val, y_preds=val_preds, s=100, color='tab20')
 
         val_list.append(val_acc)
         
@@ -501,76 +503,76 @@ def evaluate(
     real_classes = []
     predict_classes = []
     
-    try:
-        layers = ['fex']
-
-        Wc = [0] * len(W)  # Wc = Weight copy
-        true = 0
-        y_preds = []
-        acc_list = []
-        
-        for i, w in enumerate(W):
-            Wc[i] = np.copy(w)
-            
-        
-        if bar_status == True:
+    #try:
+    layers = ['fex']
 
-            test_progress = tqdm(total=len(x_test),leave=False, desc='Testing',ncols=120)
-            acc_bar = tqdm(total=1, desc="Test Accuracy", ncols=120)
-        
+    Wc = [0] * len(W)  # Wc = Weight copy
+    true = 0
+    y_preds = []
+    acc_list = []
+    
+    for i, w in enumerate(W):
+        Wc[i] = np.copy(w)
         
-        for inpIndex, Input in enumerate(x_test):
-            Input = np.array(Input)
-            Input = Input.ravel()
-            neural_layer = Input
+    
+    if bar_status == True:
 
-            for index, Layer in enumerate(layers):
+        test_progress = tqdm(total=len(x_test),leave=False, desc='Testing',ncols=120)
+        acc_bar = tqdm(total=1, desc="Test Accuracy", ncols=120)
+    
+    
+    for inpIndex, Input in enumerate(x_test):
+        Input = np.array(Input)
+        Input = Input.ravel()
+        neural_layer = Input
 
-                if Layer == 'fex':
-                    neural_layer = fex(neural_layer, W[index], False, None, activation_potentiation)
+        for index, Layer in enumerate(layers):
 
+            if Layer == 'fex':
+                neural_layer = fex(neural_layer, W[index], False, None, activation_potentiation)
 
-            for i, w in enumerate(Wc):
-                W[i] = np.copy(w)
-                
-            neural_layer = Softmax(neural_layer)
-            max_value = max(neural_layer)
 
-            predict_probabilitys.append(max_value)
-            
+        for i, w in enumerate(Wc):
+            W[i] = np.copy(w)
             
-            RealOutput = np.argmax(y_test[inpIndex])
-            real_classes.append(RealOutput)
-            PredictedOutput = np.argmax(neural_layer)
-            predict_classes.append(PredictedOutput)
+        neural_layer = Softmax(neural_layer)
+        max_value = max(neural_layer)
 
-            if RealOutput == PredictedOutput:
-                true += 1
-            acc = true / len(y_test)
+        predict_probabilitys.append(max_value)
+        
+        
+        RealOutput = np.argmax(y_test[inpIndex])
+        real_classes.append(RealOutput)
+        PredictedOutput = np.argmax(neural_layer)
+        predict_classes.append(PredictedOutput)
 
+        if RealOutput == PredictedOutput:
+            true += 1
+        acc = true / len(y_test)
 
-            acc_list.append(acc)
-            y_preds.append(PredictedOutput)
-            
-            if bar_status == True:
-                test_progress.update(1)
-                if inpIndex == 0:
-                    acc_bar.update(acc)
-                    
-                else:
-                    acc = acc - acc_list[inpIndex - 1]
-                    acc_bar.update(acc)
-            
-        if show_metrices == True:
-            plot_evaluate(x_test, y_test, y_preds, acc_list)
-        
+
+        acc_list.append(acc)
+        y_preds.append(PredictedOutput)
         
-        for i, w in enumerate(Wc):
-            W[i] = np.copy(w)
+        if bar_status == True:
+            test_progress.update(1)
+            if inpIndex == 0:
+                acc_bar.update(acc)
+                
+            else:
+                acc = acc - acc_list[inpIndex - 1]
+                acc_bar.update(acc)
         
-    except:
+    if show_metrices == True:
+        plot_evaluate(x_test, y_test, y_preds, acc_list)
     
-        print(Fore.RED + 'ERROR:' + infoTestModel + Style.RESET_ALL)
+    
+    for i, w in enumerate(Wc):
+        W[i] = np.copy(w)
+    
+#except:
+
+    #print(Fore.RED + 'ERROR:' + infoTestModel + Style.RESET_ALL)
 
     return W, y_preds, acc
 
@@ -997,21 +999,21 @@ def predict_model_ram(Input, W, scaler_params=None, activation_potentiation=None
     Wc = [0] * len(W)
     for i, w in enumerate(W):
         Wc[i] = np.copy(w)
-    try:
-        neural_layer = Input
-        neural_layer = np.array(neural_layer)
-        neural_layer = neural_layer.ravel()
-        for index, Layer in enumerate(layers):
-
-            if Layer == 'fex':
-                neural_layer = fex(neural_layer, W[index], False, None, activation_potentiation)
-            elif Layer == 'cat':
-                neural_layer = np.dot(W[index], neural_layer)
-
-    except:
-        print(Fore.RED + "ERROR: Unexpected input or wrong model parameters from: predict_model_ram." +
-              infopredict_model_ram + Style.RESET_ALL)
-        return 'e'
+    #try:
+    neural_layer = Input
+    neural_layer = np.array(neural_layer)
+    neural_layer = neural_layer.ravel()
+    for index, Layer in enumerate(layers):
+
+        if Layer == 'fex':
+            neural_layer = fex(neural_layer, W[index], False, None, activation_potentiation)
+        elif Layer == 'cat':
+            neural_layer = np.dot(W[index], neural_layer)
+
+        """    except:
+    print(Fore.RED + "ERROR: Unexpected input or wrong model parameters from: predict_model_ram." +
+            infopredict_model_ram + Style.RESET_ALL)
+    return 'e'"""
     for i, w in enumerate(Wc):
         W[i] = np.copy(w)
     return neural_layer
@@ -1423,6 +1425,7 @@ def confusion_matrix(y_true, y_pred, class_count):
 
 def plot_evaluate(x_test, y_test, y_preds, acc_list):
     
+    
     acc = acc_list[len(acc_list) - 1]
     y_true = decode_one_hot(y_test)
 
@@ -1433,11 +1436,9 @@ def plot_evaluate(x_test, y_test, y_preds, acc_list):
     precision, recall, f1 = metrics(y_test, y_preds)
     
     
-    # Confusion matrix
     cm = confusion_matrix(y_true, y_preds, len(Class))
     fig, axs = plt.subplots(2, 2, figsize=(16, 12))
-    
-    # Confusion Matrix
+
     sns.heatmap(cm, annot=True, fmt='d', ax=axs[0, 0])
     axs[0, 0].set_title("Confusion Matrix")
     axs[0, 0].set_xlabel("Predicted Class")
@@ -1445,7 +1446,7 @@ def plot_evaluate(x_test, y_test, y_preds, acc_list):
     
     if len(Class) == 2:
         fpr, tpr, thresholds = roc_curve(y_true, y_preds)
-        # ROC Curve
+   
         roc_auc = np.trapz(tpr, fpr)
         axs[1, 0].plot(fpr, tpr, color='darkorange', lw=2, label=f'ROC curve (area = {roc_auc:.2f})')
         axs[1, 0].plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')
@@ -1523,21 +1524,50 @@ def plot_evaluate(x_test, y_test, y_preds, acc_list):
     axs[0, 1].set_title('Precision, Recall, F1 Score, and Accuracy (Weighted)')
     axs[0, 1].grid(True, axis='y', linestyle='--', alpha=0.7)
     
-
+    
     if x_test.shape[1] > 2:
 
-        X_pca = pca(x_test, n_components=2)
+            X_pca = pca(x_test, n_components=2)
     else:
-        X_pca = x_test
-    
-    num_classes = len(y_test[0])
+            X_pca = x_test
+
+    y_test = decode_one_hot(y_test)
+    num_classes = len(np.unique(y_test))
     
 
     cmap = plt.get_cmap('tab20')
 
-    axs[1,1].scatter(X_pca[:, 0], X_pca[:, 1], c=y_preds, edgecolor='k', s=100, cmap=cmap, vmin=0, vmax=num_classes - 1)
-    axs[1,1].set_title("Decision Space")
-   
+
+
+    norm = plt.Normalize(vmin=0, vmax=num_classes - 1)
+    
+
+    scatter = axs[1, 1].scatter(X_pca[:, 0], X_pca[:, 1], c=y_test, edgecolor='k', s=50, cmap=cmap, norm=norm)
+    
+
+    for cls in range(num_classes):
+        class_points = []
+
+  
+        for i in range(len(y_test)):
+            if y_preds[i] == cls:
+                class_points.append(X_pca[i])
+        
+        class_points = np.array(class_points)
+
+        if len(class_points) > 2: 
+            hull = ConvexHull(class_points)
+            hull_points = class_points[hull.vertices]
+            
+
+            hull_points = np.vstack([hull_points, hull_points[0]])
+            
+
+            axs[1, 1].fill(hull_points[:, 0], hull_points[:, 1], color=cmap(norm(cls)), alpha=0.3, edgecolor='k', label=f'Class {cls} Hull')
+
+
+    axs[1, 1].set_title("Decision Boundry")
+
     plt.show()
 
 def pca(X, n_components):
@@ -1568,7 +1598,7 @@ def pca(X, n_components):
     
     return X_reduced
 
-def plot_decision_space(x, y, s=100, color='tab20'):
+def plot_decision_boundry(x, y, y_preds=None, s=100, color='tab20'):
     
     if x.shape[1] > 2:
 
@@ -1576,15 +1606,43 @@ def plot_decision_space(x, y, s=100, color='tab20'):
     else:
         X_pca = x
 
+    if y_preds == None:
+        y_preds = y
+        
 
+    y = decode_one_hot(y)
     num_classes = len(np.unique(y))
     
-
     cmap = plt.get_cmap(color)
 
 
-    plt.scatter(X_pca[:, 0], X_pca[:, 1], c=y, edgecolor='k', s=s, cmap=cmap, vmin=0, vmax=num_classes - 1)
-    plt.title("Decision Space")
+    norm = plt.Normalize(vmin=0, vmax=num_classes - 1)
+    
+
+    scatter = plt.scatter(X_pca[:, 0], X_pca[:, 1], c=y, edgecolor='k', s=50, cmap=cmap, norm=norm)
+    
+
+    for cls in range(num_classes):
+
+        class_points = []
+
+
+        for i in range(len(y)):
+            if y_preds[i] == cls:
+                class_points.append(X_pca[i])
+        
+        class_points = np.array(class_points)
+        
+
+        if len(class_points) > 2:
+            hull = ConvexHull(class_points)
+            hull_points = class_points[hull.vertices]
+
+            hull_points = np.vstack([hull_points, hull_points[0]])
+            
+            plt.fill(hull_points[:, 0], hull_points[:, 1], color=cmap(norm(cls)), alpha=0.3, edgecolor='k', label=f'Class {cls} Hull')
+
+    plt.title("Decision Boundry")
 
     plt.draw()
     

{pyerualjetwork-2.8.1.dist-info → pyerualjetwork-2.8.2.dist-info}/METADATA

@@ -1,7 +1,7 @@
 Metadata-Version: 2.1
 Name: pyerualjetwork
-Version: 2.8.1
-Summary: Code improvoments
+Version: 2.8.2
+Summary: plot_decision_boundry function added
 Home-page: UNKNOWN
 Author: Hasan Can Beydili
 Author-email: tchasancan@gmail.com

pyerualjetwork-2.8.2.dist-info/RECORD

@@ -0,0 +1,6 @@
+plan/__init__.py,sha256=SLgk4uuvhVr3CjscpcqN2TQMGyAd4Pe8YBiXfl3brd8,525
+plan/plan.py,sha256=DUNmfjy81eQBB2gtQDLxOpTio7d8i5-H3VbietxVO-Y,56026
+pyerualjetwork-2.8.2.dist-info/METADATA,sha256=YuQspdVsLubSt0LaA5vLv_tzp7Q7WzjOcvWtzK4YtP4,331
+pyerualjetwork-2.8.2.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
+pyerualjetwork-2.8.2.dist-info/top_level.txt,sha256=G0Al3HuNJ88434XneyDtRKAIUaLCizOFYFYNhd7e2OM,5
+pyerualjetwork-2.8.2.dist-info/RECORD,,

pyerualjetwork-2.8.1.dist-info/RECORD

@@ -1,6 +0,0 @@
-plan/__init__.py,sha256=WStB_zoAADCWXzIqieDZo4frPQMEZzxXnRZ6QiQKSGY,523
-plan/plan.py,sha256=x2sLhTt97E2lclfsBKFexvLEAgPUfH8uPc2hg9ggpUM,54731
-pyerualjetwork-2.8.1.dist-info/METADATA,sha256=pgJXu2Fbh7ItUXFke792-D_zvGxU7Y8JB_1XnPaLwYI,312
-pyerualjetwork-2.8.1.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
-pyerualjetwork-2.8.1.dist-info/top_level.txt,sha256=G0Al3HuNJ88434XneyDtRKAIUaLCizOFYFYNhd7e2OM,5
-pyerualjetwork-2.8.1.dist-info/RECORD,,