rishal-hurbans / grokking-artificial-intelligence-algorithms Goto Github PK

The equations for the cognitive component and social component are swapped in the summary PSO image.

Getting following error while executing ml_scikitlearn_decision_tree_classification_simple.py :

Traceback (most recent call last):
File "ml_scikitlearn_decision_tree_classification_simple.py", line 54, in
graph.write_png('tree.png')
File "C:\Installers\Python\Python37\lib\site-packages\pydotplus\graphviz.py", line 1810, in
prog=self.prog: self.write(path, format=f, prog=prog)
File "C:\Installers\Python\Python37\lib\site-packages\pydotplus\graphviz.py", line 1918, in write
fobj.write(self.create(prog, format))
File "C:\Installers\Python\Python37\lib\site-packages\pydotplus\graphviz.py", line 1960, in create
'GraphViz's executables not found')
pydotplus.graphviz.InvocationException: GraphViz's executables not found

Hello Rishal Hurbans,

Firstly, congratulations for your book, it is awesome.

I have a question regarding ml_scikitlearn_linear_regression code. x is usually used for the columns that we use for prediction and y for the labels we want to predict. However, in the ml_scikitlearn_linear_regression code, reg_test_x is used for price, and reg_test_y is used for carat (lines 26 and 27). Is it the other way around? Maybe I am misunderstanding something.

Best,
hsang

Problem:
Executing the scripts

• ml_scikitlearn_linear_regression.py
• ml_scikitlearn_decision_tree_classification.py
• ml_scikitlearn_decision_tree_classification_simple.py

leads to the import error
ModuleNotFoundError: No module named 'ml_data_preparation'

Suggestion:

• rearange folder structure (joint dependencies in subfolder) or
• add parent folder to python path

Problem:
Executing the script carnival_bruteforce.py has extremely high memory requirements such that the process is eventually killed on my system (12 GB RAM).

Suggestion:
Use the permutation iterator instead of converting it into a set.