Sturff

adjustments_investigations.py

@@ -0,0 +1,67 @@
+"""
+This files investigate some of the special thing discovered
+when writing the paper.
+
+
+TODO: DO THIS SHIT
+* Compering the Huckleburry and the prince from Twain's
+writings.
+"""
+
+
+from core import *
+import matplotlib.pyplot as plt
+
+
+def f1():
+    Novel1, Novel2 = "Adventures of Huckleberry Finn.txt", \
+                     "The Prince and The Pauper.txt"
+    Mark = Author(MARK_TWAIN, IgnoreSpecialNoun=True)
+    for Work, Matrix in zip(Mark.list_of_works(), Mark.get_matrices()):
+        print(Work)
+        if Work == Novel1:
+            Novel1 = Matrix
+        if Work == Novel2:
+            Novel2 = Matrix
+    print("The distance between \"Adventrues of Huckleberry Finn.txt\""
+          "and \"The Prince and The Pauper.txt\"is: ")
+    print(dis(Novel1, Novel2, MatrixMetric.TwoNorm))
+    print("Ignoring Special Nouns and using matrix 2 norm")
+
+def f2():
+    """
+        Comparing thespecial nouns impact on Huckleberry Finn,
+        in tm27.
+    :return:
+    """
+    A1 = Author(MARK_TWAIN, matrixfunction=get_tm27, IgnoreSpecialNoun=True)
+    A1WorkDict = A1.work_matrix_dict()
+    WorkName = "Adventures of Huckleberry Finn.txt"
+    FinnMatrix = A1WorkDict[WorkName]
+    CentroidMatrix = A1.get_center()
+
+    DifferenceMatrix = FinnMatrix - CentroidMatrix
+    plt.imshow(DifferenceMatrix)
+    plt.colorbar()
+    plt.title("Difference_Matrix")
+    plt.savefig("Difference_Matrix")
+    plt.clf()
+
+    plt.imshow(CentroidMatrix)
+    plt.colorbar()
+    plt.title("CentroidMatrix")
+    plt.savefig("CentroidMatrix")
+    plt.clf()
+
+    plt.imshow(CentroidMatrix)
+    plt.colorbar()
+    plt.title("FinnMatrix")
+    plt.savefig("FinnMatrix")
+    plt.clf()
+
+
+    pass
+
+if __name__ == "__main__":
+    f2()
+    pass

core.py

@@ -14,9 +14,9 @@
 from os import listdir
 from os.path import isfile
 
-__all__ = ["Author", "dis_between_authors", "get_tm27", "get_2ndtm","get_2ndlogarithm", 
+__all__ = ["Author", "dis_between_authors", "get_tm27", "get_2ndtm","get_2ndlogarithm",
            "CHARLES_DICKENS",
-           "MARK_TWAIN", "CentroidOption", "MatrixMetric", "AuthorMetric"]
+           "MARK_TWAIN", "CentroidOption", "MatrixMetric", "AuthorMetric", "dis"]
 
 # A list of authors' directory:
 CHARLES_DICKENS = "data/Charles Dickens"
@@ -114,6 +114,7 @@ def s(letter):
             npmatrix[i] /= s
     return npmatrix
 
+
 def get_2ndlogarithm(lines: List[str], skipSpecialNoun=False):
     """
         Takes the logarithm after counting the frequency,
@@ -182,7 +183,7 @@ class AuthorMetric(enum.Enum):
 
     AverageDis = 2 # Taking the average distance of the given transition matrix with respect to
     # All the matrices of the author.
-    # TODO: SOMETHING IS WRONG HERE.
+
 
     CentroidDis = 3 # This metric take the matrix norm on the difference of 2 centroids of the author.
 
@@ -273,6 +274,14 @@ def list_of_works(self):
     def list_of_works_content(self):
         return list(self.__FilePathToLines.values())
 
+    def work_matrix_dict(self):
+        """
+            Give a dictionary that maps the name of the works to the
+            transition matrices.
+        :return:
+        """
+        return dict(zip(self.list_of_works(), self.get_matrices()))
+
     def name(self):
         return self.__AuthorName
 
@@ -510,7 +519,6 @@ def save_nparray(m, name):
     print("Ok, we are going to save some centroid for both of the authors now: ")
 
 
-
 def save_matrices_forall_data():
     """
         This function will save all the works of the authors's transition
@@ -521,13 +529,36 @@ def save_matrices_forall_data():
         None
     """
     global Author
-    def save_listof_matrices():
-        pass
-    def gernate_all_authors():
+    def save(NpMatrix, dir:str, filename:str):
+
+        np.savetxt(fname=filename, X=NpMatrix)
+        return
+
+    def generate_all_authors():
+        AllMatrices = [get_tm27, get_2ndtm, get_2ndlogarithm]
+        FileLocations = [CHARLES_DICKENS, MARK_TWAIN]
+        ListofAuthors = []
+        for L in FileLocations:
+            for G in AllMatrices:
+                ListofAuthors.append(
+                    Author(dir=L,
+                            matrixfunction=G,
+                            IgnoreSpecialNoun=True))
+                ListofAuthors.append(
+                    Author(dir=L,
+                            matrixfunction=G,
+                            IgnoreSpecialNoun=True))
+
+        return ListofAuthors
+
+    # TODO: FINISH THIS SHIT.
+    for Aut in generate_all_authors():
+        AuthorName = Aut.name()
+        for Work, Matrix in zip(Aut.list_of_works(), Aut.get_matrices()):
+            save(Matrix, "")
         pass
 
-
-    pass # TODO: IMPLEMENT THIS SHIT
+    pass
 
 
 if __name__ == '__main__':

main.py

@@ -185,10 +185,5 @@ def print_author(A, B):
 
 
 if __name__ == "__main__":
-    # print_experiment(MatrixGeneratingFxn=get_tm27,
-    #     IgnoreSpecialNouns=True,
-    #     CentroidType=CentroidOption.AggregateMatrix,
-    #     MatrixMetric=MatrixMetric.TwoNorm,
-    #     AuthorMetric=AuthorMetric.CentroidDis)
     print_cross_compare_experiement("data/Charles Dickens", "data/Mark Twain")