pyerualjetwork 1.3.5__py3-none-any.whl → 1.3.7__py3-none-any.whl

plan/plan.py

@@ -1,10 +1,8 @@
-# -*- coding: utf-8 -*-
 """
-Created on Thu May 30 05:00:38 2024
+Created on Thu May 30 22:12:49 2024
 
-@author: hasan
+@author: hasan can beydili
 """
-
 import numpy as np
 import time
 from colorama import Fore,Style
@@ -19,13 +17,13 @@ def TrainPLAN(
     x_train: List[Union[int, float]], 
     y_train: List[Union[int, float, str]], # At least two.. and one hot encoded
     class_count: int,
-    layers = ['fex','cat'],
-    neurons = [],
-    membran_thresholds  = ['<','=='],
-    membran_potentials = [0.01,0],
-    normalizations = ['y','y'],
-    activations = ['none','none'],
-    visualize = 'n'
+    layers: List[str],
+    neurons: List[Union[int, float]],
+    membran_thresholds: List[str],
+    membran_potentials: List[Union[int, float]],
+    normalizations: List[str],
+    activations: List[str],
+    visualize: str
 ) -> str:
         
     infoPLAN = """
@@ -155,7 +153,7 @@ def TrainPLAN(
     x_train_size = len(x_train[0])
     
     W = WeightIdentification(len(layers) - 1,class_count,neurons,x_train_size)
-    Divides = SynapticDividing(class_count,W)
+    Divides, Piece = SynapticDividing(class_count,W)
     trained_W = [1] * len(W)
     print(Fore.GREEN + "Train Started with 0 ERROR" + Style.RESET_ALL,)
     train_predictions = [None] * len(y_train)
@@ -178,9 +176,11 @@ def TrainPLAN(
                     for i, ul in enumerate(Ul):
                         if ul == 1.0:
                             k = i
+
                     cs = Divides[int(k)][Windex][0]
+
        
-                    W[Windex] = SynapticPruning(w, cs, 'row', int(k),class_count)
+                    W[Windex] = SynapticPruning(w, cs, 'row', int(k),class_count,Piece[Windex],1)
 
         neural_layer = inp
         
@@ -197,9 +197,9 @@ def TrainPLAN(
                 neural_layer = Softmax(neural_layer)
                 
             if Layer == 'fex':
-                neural_layer,W[Lindex] = Fex(neural_layer, W[Lindex], membran_thresholds[Lindex], membran_potentials[Lindex])
+                neural_layer,W[Lindex] = Fex(neural_layer, W[Lindex], membran_thresholds[Lindex], membran_potentials[Lindex], Piece[Windex],1)
             elif Layer == 'cat':
-                neural_layer,W[Lindex] = Cat(neural_layer, W[Lindex], membran_thresholds[Lindex], membran_potentials[Lindex],1)
+                neural_layer,W[Lindex] = Cat(neural_layer, W[Lindex], membran_thresholds[Lindex], membran_potentials[Lindex],1, Piece[Windex])
                 
         RealOutput = np.argmax(y_train[index])
         PredictedOutput = np.argmax(neural_layer)
@@ -319,7 +319,9 @@ def SynapticPruning(
     cs,           # list[list[num]]: Synaptic connections between neurons.
     key,          # int: key for identifying synaptic connections.
     Class,        # int: Class label for the current training instance.
-    class_count   # int: Total number of classes in the dataset.
+    class_count, # int: Total number of classes in the dataset.
+    piece, # ???
+    is_training # int: 1 or 0
     
 ) -> str:
     infoPruning = """
@@ -335,18 +337,31 @@ def SynapticPruning(
     Returns:
         numpy array: Weight matrix.
     """
-
+    
     
     Class += 1 # because index start 0
     
-    if Class != class_count and Class != 1:
-            
-            ce = cs / Class
+    if  Class != 1:
+        
      
-            w[int(ce)-1::-1,:] = 0
-          
-            w[cs:,:] = 0
-    
+               
+        ce = cs / Class
+        
+        if is_training == 1:
+        
+            p = piece
+                
+            for i in range(Class - 3):
+                    
+                piece+=p
+                    
+            if Class!= 2:
+                 ce += piece
+            
+        w[int(ce)-1::-1,:] = 0
+        
+            
+        w[cs:,:] = 0
 
     else:
         
@@ -401,32 +416,36 @@ def SynapticDividing(
     # Boş bir üç boyutlu liste oluşturma
     Divides = [[[0] for _ in range(len(W))] for _ in range(class_count)]
     
+    
     for i in range(len(W)):
             
 
             Piece[i] = int(math.floor(W[i].shape[0] / class_count))
 
-    cs = 0
+    cs = 0 
     # j = Classes, i = Weights, [0] = CutStart.
 
     for i in range(len(W)):
         for j in range(class_count):
             cs = cs + Piece[i]
             Divides[j][i][0] = cs
+        #pruning_param[i] = cs
             #print('Divides: ' + j + i + ' = ' + Divides[j][i][0])
             #input()
-        
+            
         j = 0
         cs = 0
         
-    return Divides
+    return Divides, Piece
         
 
 def Fex(
     Input,               # list[num]: Input data.
     w,                   # list[list[num]]: Weight matrix of the neural network.
     membran_threshold,      # str: Sign for threshold comparison ('<', '>', '==', '!=').
-    membran_potential       # num: Threshold value for comparison.
+    membran_potential, # num: Threshold value for comparison.
+    piece, # ???
+    is_training # num: 1 or 0
 ) -> tuple:
     """
     Applies feature extraction process to the input data using synaptic pruning.
@@ -450,7 +469,7 @@ def Fex(
     elif membran_threshold == '!=':
         PruneIndex = np.where(Input != membran_potential)
 
-    w = SynapticPruning(w, PruneIndex, 'col', 0, 0)
+    w = SynapticPruning(w, PruneIndex, 'col', 0, 0, piece, is_training)
 
     neural_layer = np.dot(w, Input)
     return neural_layer,w
@@ -460,7 +479,8 @@ def Cat(
     w,                   # list[list[num]]: Weight matrix of the neural network.
     membran_threshold,      # str: Sign for threshold comparison ('<', '>', '==', '!=').
     membran_potential,      # num: Threshold value for comparison.
-    isTrain              # int: Flag indicating if the function is called during training (1 for training, 0 otherwise).
+    isTrain,
+    piece              # int: Flag indicating if the function is called during training (1 for training, 0 otherwise).
 ) -> tuple:
     """
     Applies categorization process to the input data using synaptic pruning if specified.
@@ -486,7 +506,7 @@ def Cat(
         PruneIndex = np.where(Input != membran_potential)
     if isTrain == 1 and membran_threshold != 'none':
      
-            w = SynapticPruning(w, PruneIndex, 'col', 0, 0)
+            w = SynapticPruning(w, PruneIndex, 'col', 0, 0, piece, isTrain)
      
     
     neural_layer = np.dot(w, Input)
@@ -564,6 +584,8 @@ def Relu(
     return np.maximum(0, x)
 
 
+
+
 def TestPLAN(
     x_test,         # list[list[num]]: Test input data.
     y_test,         # list[num]: Test labels.
@@ -620,9 +642,9 @@ def TestPLAN(
                         neural_layer = Softmax(neural_layer)
                         
                 if layers[index] == 'fex':
-                    neural_layer,useless = Fex(neural_layer, W[index], membran_thresholds[index], membran_potentials[index])
+                    neural_layer,useless = Fex(neural_layer, W[index], membran_thresholds[index], membran_potentials[index],0,0)
                 if layers[index] == 'cat':
-                    neural_layer,useless = Cat(neural_layer, W[index], membran_thresholds[index], membran_potentials[index],0)
+                    neural_layer,useless = Cat(neural_layer, W[index], membran_thresholds[index], membran_potentials[index],0,0)
             for i, w in enumerate(Wc):
                 W[i] = np.copy(w)
             RealOutput = np.argmax(y_test[inpIndex])
@@ -979,12 +1001,12 @@ def PredictFromDiscPLAN(Input,model_name,model_path,log_type):
             if layers[index] == 'fex':
                 neural_layer,useless = Fex(neural_layer, W[index],
                                           membran_thresholds[index],
-                                          membran_potentials[index])
+                                          membran_potentials[index],0,0)
             if layers[index] == 'cat':
                 neural_layer,useless = Cat(neural_layer, W[index],
                                           membran_thresholds[index],
                                           membran_potentials[index],
-                                          0)
+                                          0,0)
     except:
        print(Fore.RED + "ERROR: The input was probably entered incorrectly. from: PredictFromDiscPLAN"  + infoPredictFromDİscPLAN + Style.RESET_ALL)
        return 'e'
@@ -1031,11 +1053,12 @@ def PredictFromRamPLAN(Input,layers,membran_thresholds,membran_potentials,normal
             if layers[index] == 'fex':
                 neural_layer,useless = Fex(neural_layer, W[index],
                                           membran_thresholds[index],
-                                          membran_potentials[index])
+                                          membran_potentials[index],0,0)
             if layers[index] == 'cat':
                 neural_layer,useless = Cat(neural_layer, W[index],
                                           membran_thresholds[index],
-                                          membran_potentials[index],0)
+                                          membran_potentials[index],
+                                          0,0)
     except:
         print(Fore.RED + "ERROR: Unexpected input or wrong model parameters from: PredictFromRamPLAN."  + infoPredictFromRamPLAN + Style.RESET_ALL)
         return 'e'

pyerualjetwork-1.3.7.dist-info/METADATA

@@ -0,0 +1,8 @@
+Metadata-Version: 2.1
+Name: pyerualjetwork
+Version: 1.3.7
+Summary: Advanced python deep learning library. MASSIVE Technic Update, unlocked class limits. (Documentation in desc. Examples in GİTHUB: https://github.com/HCB06/PyerualJetwork)
+Author: Hasan Can Beydili
+Author-email: tchasancan@gmail.com
+Keywords: model evaluation,classifcation,pruning learning artficial neural networks
+

pyerualjetwork-1.3.7.dist-info/RECORD

@@ -0,0 +1,6 @@
+plan/__init__.py,sha256=LQbg-AnTUz7KA1E77-mg7X-zRM-7IiK7c3zK-j063rc,375
+plan/plan.py,sha256=7E8SRs8CEz6P29SM-G1VDypebXG-5oeMI2V8h0tepyU,44862
+pyerualjetwork-1.3.7.dist-info/METADATA,sha256=1Apa30bZ6idAW9qxHzSnaJAkNDJB0ADs8f3z-cfDYq0,393
+pyerualjetwork-1.3.7.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
+pyerualjetwork-1.3.7.dist-info/top_level.txt,sha256=G0Al3HuNJ88434XneyDtRKAIUaLCizOFYFYNhd7e2OM,5
+pyerualjetwork-1.3.7.dist-info/RECORD,,

pyerualjetwork-1.3.5.dist-info/METADATA

@@ -1,8 +0,0 @@
-Metadata-Version: 2.1
-Name: pyerualjetwork
-Version: 1.3.5
-Summary: Advanced python deep learning library. New Features: 'SyntheticAugmentation' function added for unbalanced datasets. Changes for variable names to snake_case (Function names are still PascalCase). (Documentation in desc. Examples in GİTHUB: https://github.com/HCB06/PyerualJetwork)
-Author: Hasan Can Beydili
-Author-email: tchasancan@gmail.com
-Keywords: model evaluation,classifcation,pruning learning artficial neural networks
-

pyerualjetwork-1.3.5.dist-info/RECORD

@@ -1,6 +0,0 @@
-plan/__init__.py,sha256=LQbg-AnTUz7KA1E77-mg7X-zRM-7IiK7c3zK-j063rc,375
-plan/plan.py,sha256=GxjRfztryI-GN_1p0dIwONgwMZW40tkslVY5N6BfjBw,44286
-pyerualjetwork-1.3.5.dist-info/METADATA,sha256=Go_Ix7kNK67fvUKKmWYZdnDW_qBl-wAgAcuCaCxbxTI,504
-pyerualjetwork-1.3.5.dist-info/WHEEL,sha256=2wepM1nk4DS4eFpYrW1TTqPcoGNfHhhO_i5m4cOimbo,92
-pyerualjetwork-1.3.5.dist-info/top_level.txt,sha256=G0Al3HuNJ88434XneyDtRKAIUaLCizOFYFYNhd7e2OM,5
-pyerualjetwork-1.3.5.dist-info/RECORD,,