Mini-Projekt 2020-03-25: Begriffsbaum mithilfe von SPARQL & Python erstellen - Teil 3

Begriffsbaum mithilfe von SPARQL & Python erstellen - Teil 3

Hui! Fast schon fertig!
(Nur ein, zwei Kleinigkeiten sind noch suboptimal - finde die Fehler!)

Code

from ete3 import Tree
import pandas as pd

#Warmspielereien
unrooted_tree = Tree( "(A,B,(C,D));" )
print(unrooted_tree)

rooted_tree = Tree( "((A,B),(C,D));" )
print(rooted_tree)

rooted_tree = Tree( "(Informatik,((Bastelstube,Bastelhuber),(Wastel,(Wastelhuber,Splock))));" )
print(rooted_tree)

rooted_tree = Tree( "(Bastelstube,Bastelhuber,(Wastelhuber,Splock)Wastl)Informatik;", format=1 )
print(rooted_tree.get_ascii(show_internal=True))

#Lesen des als pickle gespeicherten DFs
unpickled_df = pd.read_pickle("/home/zarko/Dokumente/KnowledgeGraph/subclasses_df.pickle")

print(unpickled_df)

#Programm
from ete3 import Tree, TreeStyle, TextFace, add_face_to_node
from collections import defaultdict

def build_tree(nodes):
    root = None
    for i in nodes:
        if i[0] == 'Wurzel':
            root = i
    if not root:
        raise ValueError('no root!')
    data = defaultdict(list)
    for i in nodes:
        if i[0] != 'Wurzel':
            data[i[1]] = []
            data[i[0]].append(i[1])
    return data, root[1]

def dict_tree_to_str(tree, root):
    if not tree[root]:
        return f'{root}'
    subtrees = [dict_tree_to_str(tree, st) for st in tree[root]]
    return f'({",".join(subtrees)}){root}'

#Baum-Liste bilden
tree_csv = [['Wurzel','Informatik']]
for row in unpickled_df.itertuples(index=True, name='Pandas'):
    print(row[4],row[2])
    tree_csv.append([row[4],row[2]])
print(tree_csv)

#tree_csv = [[-1, 1], [1, 2], [1, 3], [2, 4], [3, 5], [3, 6], [5, 7]]
tree_dict, root = build_tree(tree_csv)
t = Tree(dict_tree_to_str(tree_dict, root)+';',  format=1)

ts = TreeStyle()
ts.show_leaf_name = False
def my_layout(node): # https://github.com/etetoolkit/ete/issues/219
    F = TextFace(node.name, tight_text=True)
    add_face_to_node(F, node, column=0, position="branch-right")
ts.layout_fn = my_layout

t.render("/home/zarko/Dokumente/KnowledgeGraph/mytree.png", w=183, units="mm", tree_style=ts)

Output (Nur erzeugtes Bild)

Links

pandas.read_pickle — pandas 1.0.3 documentation Tree Structure in python using CSV data - Stack Overflow ete3.Tree Python Example Working With Tree Data Structures — ETE Toolkit - analysis and visualization of trees ETE Toolkit - Download ETE Toolkit - Analysis and Visualization of (phylogenetic) trees Tree plotting in Python - Stack Overflow GitHub - phrb/3d-spanning-tree: Python script for generating 3d spanning trees. [Personal project] I made 3D Trees generator in Python (youtube demo) : Python pandas.DataFrame.to_pickle — pandas 1.0.3 documentation

Vorfazit

Das kommt der Sache schon sehr nahe, die ich mit dem Herein-Hinaus-Zoomen meinen wollte (in diesem Blog-Post: https://slaughterlaughter.blogspot.com/2020/03/kompetenz.html)
Diesbezüglich - aber auch darüberhinausbezüglich - habe ich noch etwas vor. Ett kütt. Bin aber grade etwas ruhebedürftig, so wie sich das bei mir anfühlt. Corona + Coden, das strengt schon etwas an, auch ohne große Kraftanstrengung. Öh, ... äh... iwi so. lalalo.


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Kleiner Nachschub

...

ts = TreeStyle()
ts.show_leaf_name = True
ts.mode = "c"
ts.arc_start = -360 # 0 degrees = 3 o'clock
ts.arc_span = 360
t.show(tree_style=ts)

http://etetoolkit.org/docs/latest/tutorial/tutorial_drawing.html