Davout 0.1.1.dev151__tar.gz → 0.1.1.dev152__tar.gz

{davout-0.1.1.dev151/source/Davout.egg-info → davout-0.1.1.dev152}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: Davout
-Version: 0.1.1.dev151
+Version: 0.1.1.dev152
 Summary: Scientific computing utilities combining FEM, ANN and tooling
 Author: Matheus Janczkowski
 Author-email: Matheus Janczkowski <matheusj2009@hotmail.com>

{davout-0.1.1.dev151 → davout-0.1.1.dev152}/pyproject.toml

@@ -63,7 +63,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "Davout"
-version = "0.1.1.dev151"
+version = "0.1.1.dev152"
 description = "Scientific computing utilities combining FEM, ANN and tooling"
 readme = "README.md"
 authors = [

{davout-0.1.1.dev151 → davout-0.1.1.dev152}/source/Davout/DeepMech/aa_tests/tests_tool_box/test_SVD_based_NNs.py

@@ -362,6 +362,9 @@ class TestSVDArchitecture:
         number_of_neurons_hidden_layer_main_network_performance = [[500],
         [850], [1200]]
 
+        #number_of_neurons_hidden_layer_main_network_performance = [[1550],
+        #[1900], [2250]]
+
         output_dimension_performance = 2000
 
         modulating_function_performance = "identity"

{davout-0.1.1.dev151 → davout-0.1.1.dev152}/source/Davout/DeepMech/custom_architectures/SVD_quotient_space.py

@@ -1364,12 +1364,34 @@ class SVDQuotientSpace:
 
         average_raw_vector = tf.reduce_mean(raw_vector)
 
+        """
+        # Option 1
         # Concatenates the non-free component of the Householder vector
         # to the first position
 
         raw_vector = tf.math.l2_normalize(tf.concat([tf.sqrt(
         tf.math.square(average_raw_vector)+
-        self.householder_epsilon_squared)[None], raw_vector], axis=0))
+        self.householder_epsilon_squared)[None], raw_vector], axis=0))#"""
+
+        #"""
+        # Option 2
+        # Evaluates the first non-zero component squared
+
+        first_component_squared = (tf.math.square(average_raw_vector)+
+        self.householder_epsilon_squared)
+
+        # Evaluates the normalization factor
+
+        normalization_factor = tf.math.rsqrt(first_component_squared+
+        tf.reduce_sum(tf.math.square(raw_vector)))
+
+        # Recomputes the raw vector by concatenating the first non-zero
+        # component and making everything unitary
+
+        raw_vector = tf.concat([tf.sqrt(first_component_squared)[None]*
+        normalization_factor, raw_vector*normalization_factor], axis=0)
+
+        #"""
 
         # Rescales the raw vector to have unit norm, then adds the trai-
         # ling zeros and returns it
@@ -1411,10 +1433,23 @@ class SVDQuotientSpace:
         # only for the non-zero components and that the result is padded
         # to be added to the original input vector
 
+        #"""
+        # Option 1
         return input_vector+tf.pad(-(self.two*tf.einsum('bi,i->b', 
         input_vector[:,number_of_leading_zeros:], 
         householder_vector_core)[:,None]*householder_vector_core[None,:]
-        ), [[0,0], [number_of_leading_zeros,0]])
+        ), [[0,0], [number_of_leading_zeros,0]])#"""
+
+        """
+        # Option 2
+        active_input = input_vector[:,number_of_leading_zeros:]
+
+        update = -(self.two*tf.einsum('bi,i->b', active_input,
+        householder_vector_core)[:,None]*householder_vector_core[None,:])
+
+        return tf.concat([input_vector[:,:number_of_leading_zeros],
+        active_input+update], axis=1)
+        #"""
     
     # Defines a function to create a wrapper for the method that multi-
     # plies the input vector by an orthogonal matrix and, then, calls

{davout-0.1.1.dev151 → davout-0.1.1.dev152}/source/Davout/GraphUtilities/aa_tests/strain_tensors.py

@@ -3,6 +3,8 @@
 
 import matplotlib.pyplot as plt
 
+from sklearn.linear_model import LinearRegression
+
 import numpy as np
 
 # Defines a function to evaluate the rotation tensor as a function of 
@@ -108,7 +110,7 @@ if __name__=="__main__":
 
     differences = [difference_between_strain_tensors(t) for t in t_range]
 
-    plt.figure(figsize=(8, 5))
+    """plt.figure(figsize=(8, 5))
 
     plt.plot(t_range, differences, color="crimson", linewidth=2)
 
@@ -124,4 +126,18 @@ if __name__=="__main__":
 
     plt.tight_layout()
 
-    plt.show()
+    plt.show()"""
+
+    x = [500.0, 850.0, 1200.0]
+
+    y = [3.424, 5.711, 8.086]
+
+    x = np.array([[x_i] for x_i in x])
+
+    y = np.array(y)
+
+    model = LinearRegression()
+
+    model.fit(x, y)
+    
+    print("Slope: "+str(model.coef_[0]))

{davout-0.1.1.dev151 → davout-0.1.1.dev152/source/Davout.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: Davout
-Version: 0.1.1.dev151
+Version: 0.1.1.dev152
 Summary: Scientific computing utilities combining FEM, ANN and tooling
 Author: Matheus Janczkowski
 Author-email: Matheus Janczkowski <matheusj2009@hotmail.com>