pyerualjetwork 5.0.3__py3-none-any.whl → 5.2__py3-none-any.whl

pyerualjetwork/__init__.py

@@ -2,8 +2,8 @@
 
 PyerualJetwork
 ==============
-PyereualJetwork is a large, GPU-accelerated machine learning library in Python designed for professionals and researchers.
-It features PLAN, MLP, and Deep Learning training, as well as ENE (Eugenic NeuroEvolution) for genetic optimization, 
+PyereualJetwork is a large wide GPU-accelerated machine learning library in Python designed for professionals and researchers.
+It features PLAN, MLP Deep Learning and PTNN training, as well as ENE (Eugenic NeuroEvolution) for genetic optimization, 
 which can also be applied to genetic algorithms or Reinforcement Learning (RL) problems. 
 The library includes functions for data pre-processing, visualizations, model saving and loading, prediction and evaluation, 
 training, and both detailed and simplified memory management.
@@ -42,7 +42,7 @@ PyerualJetwork document: https://github.com/HCB06/PyerualJetwork/blob/main/Welco
 - Contact: tchasancan@gmail.com
 """
 
-__version__ = "5.0.3"
+__version__ = "5.2"
 __update__ = """* Changes: https://github.com/HCB06/PyerualJetwork/blob/main/CHANGES
 * PyerualJetwork Homepage: https://github.com/HCB06/PyerualJetwork/tree/main
 * PyerualJetwork document: https://github.com/HCB06/PyerualJetwork/blob/main/Welcome_to_PyerualJetwork/PYERUALJETWORK_USER_MANUEL_AND_LEGAL_INFORMATION(EN).pdf

pyerualjetwork/activation_functions_cpu.py

@@ -6,9 +6,9 @@ import warnings
 # ACTIVATION FUNCTIONS -----
 
 def all_activations():
-    
-    activations_list = ['linear', 'sigmoid', 'relu', 'tanh', 'circular', 'spiral', 'swish', 'sin_plus', 'mod_circular', 'tanh_circular', 'leaky_relu', 'softplus', 'elu', 'gelu', 'selu', 'sinakt', 'p_squared', 'sglu', 'dlrelu', 'exsig', 'acos', 'srelu', 'qelu', 'isra', 'waveakt', 'arctan', 'bent_identity', 'sech', 'softsign', 'pwl', 'cubic', 'gaussian', 'sine', 'tanh_square', 'mod_sigmoid', 'quartic', 'square_quartic', 'cubic_quadratic', 'sine_square', 'logarithmic', 'scaled_cubic', 'sine_offset']
-    
+    # softplus, cubic, square_quartic, cubic_quadratic, scaled_cubic out.
+    activations_list = ['linear', 'sigmoid', 'relu', 'tanh', 'circular', 'spiral', 'sin_plus', 'mod_circular', 'tanh_circular', 'leaky_relu', 'gelu', 'sinakt', 'p_squared', 'sglu', 'dlrelu', 'acos', 'isra', 'waveakt', 'arctan', 'bent_identity', 'softsign', 'pwl', 'sine', 'tanh_square', 'sine_square', 'logarithmic', 'sine_offset']
+    # suggest: tanh, arctan, selu, elu, dlrelu, isra, bent_identity
     return activations_list
 
 def spiral_activation(x):
@@ -76,9 +76,6 @@ def Relu(
 def tanh(x):
     return np.tanh(x)
 
-def swish(x):
-    return x * (1 / (1 + np.exp(-x)))
-
 def sin_plus(x):
     return (np.sin(x) + 1) / 2
 
@@ -91,18 +88,9 @@ def tanh_circular_activation(x):
 def leaky_relu(x, alpha=0.01):
     return np.where(x > 0, x, alpha * x)
 
-def softplus(x):
-    return np.log(1 + np.exp(x))
-
-def elu(x, alpha=1.0):
-    return np.where(x > 0, x, alpha * (np.exp(x) - 1))
-
 def gelu(x):
     return 0.5 * x * (1 + np.tanh(np.sqrt(2 / np.pi) * (x + 0.044715 * np.power(x, 3))))
 
-def selu(x, lambda_=1.0507, alpha=1.6733):
-    return lambda_ * np.where(x > 0, x, alpha * (np.exp(x) - 1))
-
 def sinakt(x):
     return np.sin(x) + np.cos(x)
 
@@ -116,26 +104,10 @@ def sglu(x, alpha=1.0):
 def dlrelu(x):
     return np.maximum(0.01 * x, x) + np.minimum(0.01 * x, 0.1 * x)
 
-# 5. Exponential Sigmoid (ExSig)
-def exsig(x):
-    return 1 / (1 + np.exp(-x**2))
-
 # 6. Adaptive Cosine Activation (ACos)
 def acos(x, alpha=1.0, beta=0.0):
     return np.cos(alpha * x + beta)
 
-# 7. Gaussian-like Activation (GLA)
-def gla(x, alpha=1.0, mu=0.0):
-    return np.exp(-alpha * (x - mu)**2)
-
-# 8. Swish ReLU (SReLU)
-def srelu(x):
-    return x * (1 / (1 + np.exp(-x))) + np.maximum(0, x)
-
-# 9. Quadratic Exponential Linear Unit (QELU)
-def qelu(x):
-    return x**2 * np.exp(x) - 1
-
 # 10. Inverse Square Root Activation (ISRA)
 def isra(x):
     return x / np.sqrt(np.abs(x) + 1)
@@ -170,48 +142,24 @@ def circular_activation(x, scale=2.0, frequency=1.0, shift=0.0):
     
     return circular_output
 
-def sech(x):
-    return 2 / (np.exp(x) + np.exp(-x))
-
 def softsign(x):
     return x / (1 + np.abs(x))
 
 def pwl(x, alpha=0.5, beta=1.5):
     return np.where(x <= 0, alpha * x, beta * x)
 
-def cubic(x):
-    return x**3
-
-def gaussian(x, alpha=1.0, mu=0.0):
-    return np.exp(-alpha * (x - mu)**2)
-                
 def sine(x, alpha=1.0):
     return np.sin(alpha * x)
 
 def tanh_square(x):
     return np.tanh(x)**2
 
-def mod_sigmoid(x, alpha=1.0, beta=0.0):
-    return 1 / (1 + np.exp(-alpha * x + beta))
-
-def quartic(x):
-    return x**4
-
-def square_quartic(x):
-    return (x**2)**2
-
-def cubic_quadratic(x):
-    return x**3 * (x**2)
-
 def sine_square(x):
     return np.sin(x)**2
 
 def logarithmic(x):
     return np.log(x**2 + 1)
 
-def scaled_cubic(x, alpha=1.0):
-    return alpha * x**3
-
 def sine_offset(x, beta=0.0):
     return np.sin(x + beta)
 
@@ -228,45 +176,29 @@ def apply_activation(Input, activation_list):
     
     activation_functions = {
         'sigmoid': Sigmoid,
-        'swish': swish,
         'mod_circular': modular_circular_activation,
         'tanh_circular': tanh_circular_activation,
         'leaky_relu': leaky_relu,
         'relu': Relu,
-        'softplus': softplus,
-        'elu': elu,
         'gelu': gelu,
-        'selu': selu,
         'tanh': tanh,
         'sinakt': sinakt,
         'p_squared': p_squared,
         'sglu': lambda x: sglu(x, alpha=1.0),
         'dlrelu': dlrelu,
-        'exsig': exsig,
         'sin_plus': sin_plus,
         'acos': lambda x: acos(x, alpha=1.0, beta=0.0),
-        'gla': lambda x: gla(x, alpha=1.0, mu=0.0),
-        'srelu': srelu,
-        'qelu': qelu,
         'isra': isra,
         'waveakt': waveakt,
         'arctan': arctan,
         'bent_identity': bent_identity,
-        'sech': sech,
         'softsign': softsign,
         'pwl': pwl,
-        'cubic': cubic,
-        'gaussian': gaussian,
         'sine': sine,
         'tanh_square': tanh_square,
-        'mod_sigmoid': mod_sigmoid,
         'linear': lambda x: x,
-        'quartic': quartic,
-        'square_quartic': square_quartic,
-        'cubic_quadratic': cubic_quadratic,
         'sine_square': sine_square,
         'logarithmic': logarithmic,
-        'scaled_cubic': lambda x: scaled_cubic(x, 1.0),
         'sine_offset': lambda x: sine_offset(x, 1.0),
         'spiral': spiral_activation,
         'circular': circular_activation

pyerualjetwork/activation_functions_cuda.py

@@ -6,7 +6,7 @@ import warnings
 
 def all_activations():
     
-    activations_list = ['linear', 'sigmoid', 'relu', 'tanh', 'circular', 'spiral', 'swish', 'sin_plus', 'mod_circular', 'tanh_circular', 'leaky_relu', 'softplus', 'elu', 'gelu', 'selu', 'sinakt', 'p_squared', 'sglu', 'dlrelu', 'exsig', 'acos', 'srelu', 'qelu', 'isra', 'waveakt', 'arctan', 'bent_identity', 'sech', 'softsign', 'pwl', 'cubic', 'gaussian', 'sine', 'tanh_square', 'mod_sigmoid', 'quartic', 'square_quartic', 'cubic_quadratic', 'sine_square', 'logarithmic', 'scaled_cubic', 'sine_offset']
+    activations_list = ['linear', 'sigmoid', 'relu', 'tanh', 'circular', 'spiral', 'sin_plus', 'mod_circular', 'tanh_circular', 'leaky_relu', 'gelu', 'sinakt', 'p_squared', 'sglu', 'dlrelu', 'acos', 'isra', 'waveakt', 'arctan', 'bent_identity', 'softsign', 'pwl', 'sine', 'tanh_square', 'sine_square', 'logarithmic', 'sine_offset']
     
     return activations_list
 
@@ -75,9 +75,6 @@ def Relu(
 def tanh(x):
     return cp.tanh(x)
 
-def swish(x):
-    return x * (1 / (1 + cp.exp(-x)))
-
 def sin_plus(x):
     return (cp.sin(x) + 1) / 2
 
@@ -90,18 +87,6 @@ def tanh_circular_activation(x):
 def leaky_relu(x, alpha=0.01):
     return cp.where(x > 0, x, alpha * x)
 
-def softplus(x):
-    return cp.log(1 + cp.exp(x))
-
-def elu(x, alpha=1.0):
-    return cp.where(x > 0, x, alpha * (cp.exp(x) - 1))
-
-def gelu(x):
-    return 0.5 * x * (1 + cp.tanh(cp.sqrt(2 / cp.pi) * (x + 0.044715 * cp.power(x, 3))))
-
-def selu(x, lambda_=1.0507, alpha=1.6733):
-    return lambda_ * cp.where(x > 0, x, alpha * (cp.exp(x) - 1))
-
 def sinakt(x):
     return cp.sin(x) + cp.cos(x)
 
@@ -115,26 +100,10 @@ def sglu(x, alpha=1.0):
 def dlrelu(x):
     return cp.maximum(0.01 * x, x) + cp.minimum(0.01 * x, 0.1 * x)
 
-# 5. Exponential Sigmoid (ExSig)
-def exsig(x):
-    return 1 / (1 + cp.exp(-x**2))
-
 # 6. Adaptive Cosine Activation (ACos)
 def acos(x, alpha=1.0, beta=0.0):
     return cp.cos(alpha * x + beta)
 
-# 7. Gaussian-like Activation (GLA)
-def gla(x, alpha=1.0, mu=0.0):
-    return cp.exp(-alpha * (x - mu)**2)
-
-# 8. Swish ReLU (SReLU)
-def srelu(x):
-    return x * (1 / (1 + cp.exp(-x))) + cp.maximum(0, x)
-
-# 9. Quadratic Exponential Linear Unit (QELU)
-def qelu(x):
-    return x**2 * cp.exp(x) - 1
-
 # 10. Inverse Square Root Activation (ISRA)
 def isra(x):
     return x / cp.sqrt(cp.abs(x) + 1)
@@ -169,48 +138,24 @@ def circular_activation(x, scale=2.0, frequency=1.0, shift=0.0):
     
     return circular_output
 
-def sech(x):
-    return 2 / (cp.exp(x) + cp.exp(-x))
-
 def softsign(x):
     return x / (1 + cp.abs(x))
 
 def pwl(x, alpha=0.5, beta=1.5):
     return cp.where(x <= 0, alpha * x, beta * x)
 
-def cubic(x):
-    return x**3
-
-def gaussian(x, alpha=1.0, mu=0.0):
-    return cp.exp(-alpha * (x - mu)**2)
-                
 def sine(x, alpha=1.0):
     return cp.sin(alpha * x)
 
 def tanh_square(x):
     return cp.tanh(x)**2
 
-def mod_sigmoid(x, alpha=1.0, beta=0.0):
-    return 1 / (1 + cp.exp(-alpha * x + beta))
-
-def quartic(x):
-    return x**4
-
-def square_quartic(x):
-    return (x**2)**2
-
-def cubic_quadratic(x):
-    return x**3 * (x**2)
-
 def sine_square(x):
     return cp.sin(x)**2
 
 def logarithmic(x):
     return cp.log(x**2 + 1)
 
-def scaled_cubic(x, alpha=1.0):
-    return alpha * x**3
-
 def sine_offset(x, beta=0.0):
     return cp.sin(x + beta)
 
@@ -227,45 +172,28 @@ def apply_activation(Input, activation_list):
     
     activation_functions = {
         'sigmoid': Sigmoid,
-        'swish': swish,
         'mod_circular': modular_circular_activation,
         'tanh_circular': tanh_circular_activation,
         'leaky_relu': leaky_relu,
         'relu': Relu,
-        'softplus': softplus,
-        'elu': elu,
-        'gelu': gelu,
-        'selu': selu,
         'tanh': tanh,
         'sinakt': sinakt,
         'p_squared': p_squared,
         'sglu': lambda x: sglu(x, alpha=1.0),
         'dlrelu': dlrelu,
-        'exsig': exsig,
         'sin_plus': sin_plus,
         'acos': lambda x: acos(x, alpha=1.0, beta=0.0),
-        'gla': lambda x: gla(x, alpha=1.0, mu=0.0),
-        'srelu': srelu,
-        'qelu': qelu,
         'isra': isra,
         'waveakt': waveakt,
         'arctan': arctan,
         'bent_identity': bent_identity,
-        'sech': sech,
         'softsign': softsign,
         'pwl': pwl,
-        'cubic': cubic,
-        'gaussian': gaussian,
         'sine': sine,
         'tanh_square': tanh_square,
-        'mod_sigmoid': mod_sigmoid,
         'linear': lambda x: x,
-        'quartic': quartic,
-        'square_quartic': square_quartic,
-        'cubic_quadratic': cubic_quadratic,
         'sine_square': sine_square,
         'logarithmic': logarithmic,
-        'scaled_cubic': lambda x: scaled_cubic(x, 1.0),
         'sine_offset': lambda x: sine_offset(x, 1.0),
         'spiral': spiral_activation,
         'circular': circular_activation
@@ -282,4 +210,4 @@ def apply_activation(Input, activation_list):
         
     except Exception as e:
         warnings.warn(f"Error in activation processing: {str(e)}", RuntimeWarning)
-        return Input
+        return Input

pyerualjetwork/data_operations_cpu.py

@@ -89,7 +89,7 @@ def decode_one_hot(encoded_data):
     else: return np.argmax(encoded_data, axis=1)
 
 
-def split(X, y, test_size, random_state=42, dtype=np.float32):
+def split(X, y, test_size, random_state=42):
     """
     Splits the given X (features) and y (labels) data into training and testing subsets.
 
@@ -101,9 +101,7 @@ def split(X, y, test_size, random_state=42, dtype=np.float32):
         test_size (float or int): Proportion or number of samples for the test subset.
         
         random_state (int or None): Seed for random state. Default: 42.
-        
-        dtype (numpy.dtype): Data type for the arrays. np.float32 by default. Example: np.float64 or np.float16. [fp32 for balanced devices, fp64 for strong devices, fp16 for weak devices: not reccomended!]
-
+   
     Returns:
         tuple: x_train, x_test, y_train, y_test as ordered training and testing data subsets.
     """
@@ -288,7 +286,7 @@ def synthetic_augmentation(x, y, dtype=np.float32):
     Args:
         x_train: numpy array format
 
-        y_train: numpy array format
+        y_train: numpy array format (one-hot encoded)
         
         dtype (numpy.dtype): Data type for the arrays. cp.float32 by default. Example: cp.float64 or cp.float16.
         

pyerualjetwork/ene_cuda.py

@@ -902,7 +902,7 @@ def mutation(weight,
          
          if potential_activation_delete_prob > activation_delete_prob and len(activations) > 1:
             
-            random_index = random.randint(0, len(activations))
+            random_index = random.randint(0, len(activations)-1)
             activations.pop(random_index)
             
 
@@ -910,7 +910,7 @@ def mutation(weight,
 
             try:
                
-               random_index_all_act = int(random.uniform(0, len(all_acts)))
+               random_index_all_act = int(random.uniform(0, len(all_acts)-1))
                activations.append(all_acts[random_index_all_act])
 
             except:
@@ -919,12 +919,12 @@ def mutation(weight,
                activations = []
 
                activations.append(activation)
-               activations.append(all_acts[int(random.uniform(0, len(all_acts)))])
+               activations.append(all_acts[int(random.uniform(0, len(all_acts)-1))])
             
          if potential_activation_change_prob > activation_change_prob:
             
-            random_index_all_act = int(random.uniform(0, len(all_acts)))
-            random_index_genom_act = int(random.uniform(0, len(activations)))
+            random_index_all_act = int(random.uniform(0, len(all_acts)-1))
+            random_index_genom_act = int(random.uniform(0, len(activations)-1))
 
             activations[random_index_genom_act] = all_acts[random_index_all_act]
 

pyerualjetwork/fitness_functions.py

@@ -14,4 +14,3 @@ def wals(acc, loss, acc_impact, loss_impact):
     loss_impact += np.finfo(float).eps
     
     return (acc * acc_impact) + ((loss_impact / loss) * loss_impact)
-import __init__

pyerualjetwork/issue_solver.py

@@ -49,7 +49,7 @@ def update_model_to_v5(model_name, model_path, is_cuda):
                                             save_model)
     else:
 
-        from .model_operations_cpu import (get_act, 
+        from .model_operations_cpu import (get_act,
                                             get_weights,
                                             get_scaler, 
                                             get_acc,