I'm looking at the documentation here to try and manipulate the way the import statement works. My code uses imports in all forms
import <module>
import <package.module>
from <package> import <module>
from <package.module> import <function>
from <package.module> import *
My goal is: for a certain folder, let's call it myfolder, any import for any module within myfolder (however deep in the structure) should have some preprocessing. No matter how it's imported. Preprocessing in this case is to download the python file from an internal CMS and use that instead of the one on the disk.
I understand the meta_path and path_hooks part, and I think I need to work with the path_hooks to return a FileFinder object to the built-in meta's PathFinder. Here's what I have so far:
import os, sys
class PathhookOverride():
def __init__(self, path) -> None:
"""
This will be called when PathFinder() iterates through sys.path_hooks
"""
relative_path = os.path.relpath(path, os.getcwd())
if not relative_path.startswith('myfolder'):
## We want to override only imports that have myfolder as the first part of the relative path
raise ImportError
if os.path.isdir(path):
## We know that this is a directory, we don't want to handle this
print(f'PathhookOverride: {path} is a directory')
raise ImportError
dot_separated_path = ".".join(relative_path.split(os.path.sep))
print(dot_separated_path)
## Pull file here later
cache = sys.path_importer_cache
raise ImportError ## Go to next path_hook
def change_importer():
"""Inserts the finder into the import machinery"""
sys.path_hooks.insert(0, PathhookOverride)
from myfolder.package.module import function
Expected output:
When I import my module or function using any of the above formats, I should get the path of the file being imported.
i.e., in the code snippet above, it should print the dot_separated_path:
myfolder.package.module
Actual output:
PathhookOverride: c:\test1\myfolder is a directory
PathhookOverride: c:\test1\myfolder\package is a directory
The override only catches the directories. The path of the files are never sent to the override hook.
What am I missing? Thanks.
Currently I am extracting files to the temp directory of the operating system. One of the files is a Python file containing a class which I need to get a handle of. The Python's file is known, but the name of the class inside the file is unknown. But it is safe to assume, that the there is only one single class, and that the class is a subclass of another.
I tried to work with importlib, but I am not able to get a handle of the class.
So far I tried:
# Assume
# module_name contains the name of the class and -> "MyClass"
# path_module contains the path to the python file -> "../Module.py"
spec = spec_from_file_location(module_name, path_module)
module = module_from_spec(spec)
for pair in inspect.getmembers(module):
print(f"{pair[1]} is class: {inspect.isclass(pair[1])}")
When I iterate over the members of the module, none of them get printed as a class.
My class in this case is called BasicModel and the Output on the console looks like this:
BasicModel is class: False
What is the correct approach to this?
Edit:
As the content of the file was requested, here you go:
class BasicModel(Sequential):
def __init__(self, class_count: int, input_shape: tuple):
Sequential.__init__(self)
self.add(Input(shape=input_shape))
self.add(Flatten())
self.add(Dense(128, activation=nn.relu))
self.add(Dense(128, activation=nn.relu))
self.add(Dense(class_count, activation=nn.softmax))
Use dir() to get the attributes of the file and inspect to check if the attribute is a class. If so, you can create an object.
Assuming that your file's path is /tmp/mysterious you can do this:
import importlib
import inspect
from pathlib import Path
import sys
path_pyfile = Path('/tmp/mysterious.py')
sys.path.append(str(path_pyfile.parent))
mysterious = importlib.import_module(path_pyfile.stem)
for name_local in dir(mysterious):
if inspect.isclass(getattr(mysterious, name_local)):
print(f'{name_local} is a class')
MysteriousClass = getattr(mysterious, name_local)
mysterious_object = MysteriousClass()
File "C:\Users\Administrator\Documents\Mibot\oops\blinkserv.py", line 82, in __init__
self.serv = socket(AF_INET,SOCK_STREAM)
TypeError: 'module' object is not callable
Why am I getting this error?
I'm confused.
How can I solve this error?
socket is a module, containing the class socket.
You need to do socket.socket(...) or from socket import socket:
>>> import socket
>>> socket
<module 'socket' from 'C:\Python27\lib\socket.pyc'>
>>> socket.socket
<class 'socket._socketobject'>
>>>
>>> from socket import socket
>>> socket
<class 'socket._socketobject'>
This is what the error message means:
It says module object is not callable, because your code is calling a module object. A module object is the type of thing you get when you import a module. What you were trying to do is to call a class object within the module object that happens to have the same name as the module that contains it.
Here is a way to logically break down this sort of error:
"module object is not callable. Python is telling me my code trying to call something that cannot be called. What is my code trying to call?"
"The code is trying to call on socket. That should be callable! Is the variable socket is what I think it is?`
I should print out what socket is and check print(socket)
Assume that the content of YourClass.py is:
class YourClass:
# ......
If you use:
from YourClassParentDir import YourClass # means YourClass.py
In this way, you will get TypeError: 'module' object is not callable if you then tried to call YourClass().
But, if you use:
from YourClassParentDir.YourClass import YourClass # means Class YourClass
or use YourClass.YourClass(), it works.
Add to the main __init__.py in YourClassParentDir, e.g.:
from .YourClass import YourClass
Then, you will have an instance of your class ready when you import it into another script:
from YourClassParentDir import YourClass
Short answer: You are calling a file/directory as a function instead of real function
Read on:
This kind of error happens when you import module thinking it as function and call it.
So in python module is a .py file. Packages(directories) can also be considered as modules.
Let's say I have a create.py file. In that file I have a function like this:
#inside create.py
def create():
pass
Now, in another code file if I do like this:
#inside main.py file
import create
create() #here create refers to create.py , so create.create() would work here
It gives this error as am calling the create.py file as a function.
so I gotta do this:
from create import create
create() #now it works.
Here is another gotcha, that took me awhile to see even after reading these posts. I was setting up a script to call my python bin scripts. I was getting the module not callable too.
My zig was that I was doing the following:
from mypackage.bin import myscript
...
myscript(...)
when my zag needed to do the following:
from mypackage.bin.myscript import myscript
...
myscript(...)
In summary, double check your package and module nesting.
What I am trying to do is have a scripts directory that does not have the *.py extension, and still have the 'bin' modules to be in mypackage/bin and these have my *.py extension. I am new to packaging, and trying to follow the standards as I am interpreting them. So, I have at the setup root:
setup.py
scripts/
script1
mypackage/
bin/
script1.py
subpackage1/
subpackage_etc/
If this is not compliant with standard, please let me know.
It seems like what you've done is imported the socket module as import socket. Therefore socket is the module. You either need to change that line to self.serv = socket.socket(socket.AF_INET, socket.SOCK_STREAM), as well as every other use of the socket module, or change the import statement to from socket import socket.
Or you've got an import socket after your from socket import *:
>>> from socket import *
>>> serv = socket(AF_INET,SOCK_STREAM)
>>> import socket
>>> serv = socket(AF_INET,SOCK_STREAM)
Traceback (most recent call last):
File "<input>", line 1, in <module>
TypeError: 'module' object is not callable
I know this thread is a year old, but the real problem is in your working directory.
I believe that the working directory is C:\Users\Administrator\Documents\Mibot\oops\. Please check for the file named socket.py in this directory. Once you find it, rename or move it. When you import socket, socket.py from the current directory is used instead of the socket.py from Python's directory. Hope this helped. :)
Note: Never use the file names from Python's directory to save your program's file name; it will conflict with your program(s).
When configuring an console_scripts entrypoint in setup.py I found this issue existed when the endpoint was a module or package rather than a function within the module.
Traceback (most recent call last):
File "/Users/ubuntu/.virtualenvs/virtualenv/bin/mycli", line 11, in <module>
load_entry_point('my-package', 'console_scripts', 'mycli')()
TypeError: 'module' object is not callable
For example
from setuptools import setup
setup (
# ...
entry_points = {
'console_scripts': [mycli=package.module.submodule]
},
# ...
)
Should have been
from setuptools import setup
setup (
# ...
entry_points = {
'console_scripts': [mycli=package.module.submodule:main]
},
# ...
)
So that it would refer to a callable function rather than the module itself. It seems to make no difference if the module has a if __name__ == '__main__': block. This will not make the module callable.
I faced the same problem. then I tried not using
from YourClass import YourClass
I just copied the whole code of YourClass.py and run it on the main code (or current code).it solved the error
you are using the name of a module instead of the name of the class
use
import socket
and then
socket.socket(...)
its a weird thing with the module, but you can also use something like
import socket as sock
and then use
sock.socket(...)
I guess you have overridden the builtin function/variable or something else "module" by setting the global variable "module". just print the module see whats in it.
Here's a possible extra edge case that I stumbled upon and was puzzled by for a while, hope it helps someone:
In some_module/a.py:
def a():
pass
In some_module/b.py:
from . import a
def b():
a()
In some_module/__init__.py:
from .b import b
from .a import a
main.py:
from some_module import b
b()
Then because when main.py loads b, it goes via __init__.py which tries to load b.py before a.py. This means when b.py tries to load a it gets the module rather than the function - meaning you'll get the error message module object is not callable
The solution here is to swap the order in some_module/__init__.py:
from .a import a
from .b import b
Or, if this would create a circular dependency, change your file names to not match the functions, and load directly from the files rather than relying on __init__.py
I got the same error below:
TypeError: 'module' object is not callable
When calling time() to print as shown below:
import time
print(time()) # Here
So, I called time.time() as shown below:
import time
print(time.time()) # Here
Or, I imported time from time as shown below:
from time import time # Here
print(time())
Then, the error was solved:
1671301094.5742612
I had this error when I was trying to use optuna (a library for hyperparameter tuning) with LightGBM. After an hour struggle I realized that I was importing class directly and that was creating an issue.
import lightgbm as lgb
def LGB_Objective(trial):
parameters = {
'objective_type': 'regression',
'max_depth': trial.suggest_int('max_depth', 10, 60),
'boosting': trial.suggest_categorical('boosting', ['gbdt', 'rf', 'dart']),
'data_sample_strategy': 'bagging',
'num_iterations': trial.suggest_int('num_iterations', 50, 250),
'learning_rate': trial.suggest_float('learning_rate', 0.01, 1.0),
'reg_alpha': trial.suggest_float('reg_alpha', 0.01, 1.0),
'reg_lambda': trial.suggest_float('reg_lambda', 0.01, 1.0)
}
'''.....LightGBM model....'''
model_lgb = lgb(**parameters)
model_lgb.fit(x_train, y_train)
y_pred = model_lgb.predict(x_test)
return mse(y_test, y_pred, squared=True)
study_lgb = optuna.create_study(direction='minimize', study_name='lgb_regression')
study_lgb.optimize(LGB_Objective, n_trials=200)
Here, the line model_lgb = lgb(**parameters) was trying to call the cLass itself.
When I digged into the __init__.py file in site_packages folder of LGB installation as below, I identified the module which was fit to me (I was working on regression problem). I therefore imported LGBMRegressor and replaced lgb in my code with LGBMRegressor and it started working.
You can check in your code if you are importing the entire class/directory (by mistake) or the target module within the class.
from lightgbm import LGBMRegressor
def LGB_Objective(trial):
parameters = {
'objective_type': 'regression',
'max_depth': trial.suggest_int('max_depth', 10, 60),
'boosting': trial.suggest_categorical('boosting', ['gbdt', 'rf', 'dart']),
'data_sample_strategy': 'bagging',
'num_iterations': trial.suggest_int('num_iterations', 50, 250),
'learning_rate': trial.suggest_float('learning_rate', 0.01, 1.0),
'reg_alpha': trial.suggest_float('reg_alpha', 0.01, 1.0),
'reg_lambda': trial.suggest_float('reg_lambda', 0.01, 1.0)
}
'''.....LightGBM model....'''
model_lgb = LGBMRegressor(**parameters) #here I've changed lgb to LGBMRegressor
model_lgb.fit(x_train, y_train)
y_pred = model_lgb.predict(x_test)
return mse(y_test, y_pred, squared=True)
study_lgb = optuna.create_study(direction='minimize', study_name='lgb_regression')
study_lgb.optimize(LGB_Objective, n_trials=200)
A simple way to solve this problem is export thePYTHONPATH variable enviroment. For example, for Python 2.6 in Debian/GNU Linux:
export PYTHONPATH=/usr/lib/python2.6`
In other operating systems, you would first find the location of this module or the socket.py file.
check the import statements since a module is not callable.
In Python, everything (including functions, methods, modules, classes etc.) is an object.
I would like to import all methods from a module with altered names.
For instance, instead of
from module import repetitive_methodA as methodA, \
repetitive_Class1 as Class1, \
repetitive_instance4 as instance4
I'd prefer something along the lines of
from module import * as *-without-"repetitive_"
this is a rephrasing of this clumsy unanswered question, I have not been able to find a solution or similar questions yet.
You can do it this way:
import module
import inspect
for (k,v) in inspect.getmembers(module):
if k.startswith('repetitive_'):
globals()[k.partition("_")[2]] = v
Edit in response to the comment "how is this answer intended to be used?"
Suppose module looks like this:
# module
def repetitive_A():
print ("This is repetitive_A")
def repetitive_B():
print ("This is repetitive_B")
Then after running the rename loop, this code:
A()
B()
produces this output:
This is repetitive_A
This is repetitive_B
What I would do, creating a work-around...
Including you have a file named some_file.py in the current directory, which is composed of...
# some_file.py
def rep_a():
return 1
def rep_b():
return 2
def rep_c():
return 3
When you import something, you create an object on which you call methods. These methods are the classes, variables, functions of your file.
In order to get what you want, I thought It'd be a great idea to just add a new object, containing the original functions you wanted to rename. The function redirect_function() takes an object as first parameter, and will iterate through the methods (in short, which are the functions of your file) of this object : it will, then, create another object which will contain the pointer of the function you wanted to rename at first.
tl;dr : this function will create another object which contains the original function, but the original name of the function will also remain.
See example below. :)
def redirect_function(file_import, suffixe = 'rep_'):
# Lists your functions and method of your file import.
objects = dir(file_import)
for index in range(len(objects)):
# If it begins with the suffixe, create another object that contains our original function.
if objects[index][0:len(suffixe)] == suffixe:
func = eval("file_import.{}".format(objects[index]))
setattr(file_import, objects[index][len(suffixe):], func)
if __name__ == '__main__':
import some_file
redirect_function(some_file)
print some_file.rep_a(), some_file.rep_b(), some_file.rep_c()
print some_file.a(), some_file.b(), some_file.c()
This outputs...
1 2 3
1 2 3