Reputation: 1019
I tried running the following code:
from imblearn import under_sampling, over_sampling
from imblearn.over_sampling import SMOTE
sm = SMOTE(random_state=12, ratio = 1.0)
x_SMOTE, y_SMOTE = sm.fit_sample(X, y)
which gives me the error message:
ModuleNotFoundError: No module named 'imblearn'
I have tried installing the imblearn module in multiple ways, they all seem to work (there are no errors given during the installation but when I run the above code, I get an error message).
I tried istalling imblearn using the following suggested in other stackoverflow questions:
pip install -U imbalanced-learn
pip install imblearn
!pip install imblearn
pip install -c glemaitre imbalanced-learn
pip install imblearn==0.0
None of these seem to help... Any ideas? Thank you!
Upvotes: 23
Views: 125028
Reputation: 1
pip install -U imbalanced-learn
or open VSCode terminal
python -m pip install imbalanced-learn
Upvotes: 0
Reputation:
It is working for me this way
pip install imblearn
And i import this way :
from imblearn.combine import SMOTETomek
Upvotes: 0
Reputation: 11
I had the same issue which was rectified by using:
!pip install -U imbalanced-learn
Then this:
conda install -c conda-forge imbalanced-learn
Updated my conda:
conda update -n base -c conda-forge conda
Restarted the kernel.
Upvotes: 1
Reputation: 31
If you are still experiencing error after installing imblearn either on Anaconda terminal while using Vscode. Try to restart Vscode for it to reflect.
Upvotes: 0
Reputation: 11
I have been fixed it by applying the following inside a Jupyter Notebook.
!pip install imbalanced-learn==0.6.0
!pip install scikit-learn==0.22.1
Upvotes: 1
Reputation: 93
This worked for me
pip install -U imbalanced-learn
conda install -c conda-forge imbalanced-learn
Upvotes: 5
Reputation: 21
I was dealing with the same problem. Updating packages, upgrading pip or python version did not resolve the problem for me.
The issue was that pip installed package to one folder, but my jupyter notebook imported packages from another folder. To get the path from where your packages are imported, you may use:
import site
site.getsitepackages()
# /your/path/from/python
Then you may check in terminal where pip installs your packages :
pip show imblearn
If the paths do not coincide, you may manually set the path for pip in terminal:
pip config set global.target /your/path/from/python
And install your package again by
pip install imblearn
Upvotes: 2
Reputation: 133
On AWS SageMaker, follow the documentation:
!pip install imbalanced-learn
in a notebook cell.
Upvotes: 12
Reputation: 186
Those who have permission issue or failed to install it can follow this.
Upvotes: 1
Reputation: 21
Open anaconda prompt and install below module:
conda install -c conda-forge imbalanced-learn
conda install -c conda-forge/label/gcc7 imbalanced-learn
conda install -c conda-forge/label/cf201901 imbalanced-learn
Upvotes: 2
Reputation: 2456
I've come across the same problem a few days ago - trying to use imblearn
inside a Jupyter Notebook. This question led me to the solution:
conda install -c glemaitre imbalanced-learn
Notice, one of the commands you tried (pip install -c glemaitre imbalanced-learn
) doesn't make sense: -c glemaitre
is an argument for Anaconda python distributions, which tells conda
(Anaconda's CLI) to download the module from a source different than the defaults (glemaitre's channel). Since that argument is conda
-specific, it doen't apply to pip
commands.
Upvotes: 0
Reputation: 3085
I installed the module named imblearn
using anaconda command prompt.
conda install -c conda-forge imbalanced-learn
Then imported the packages
from imblearn import under_sampling, over_sampling
from imblearn.over_sampling import SMOTE
Again, I tried to install imblearn
through pip, it works for me.
(base) C:\WINDOWS\system32>pip install -U imbalanced-learn
Requirement already up-to-date: imbalanced-learn in c:\users\ashok\anaconda3\lib\site-packages (0.4.3)
Requirement already satisfied, skipping upgrade: numpy>=1.8.2 in c:\users\ashok\anaconda3\lib\site-packages (from imbalanced-learn) (1.15.3)
Requirement already satisfied, skipping upgrade: scipy>=0.13.3 in c:\users\ashok\anaconda3\lib\site-packages (from imbalanced-learn) (0.19.1)
Requirement already satisfied, skipping upgrade: scikit-learn>=0.20 in c:\users\ashok\anaconda3\lib\site-packages (from imbalanced-learn) (0.20.0)
Upvotes: 24
Reputation: 461
try this way:
from imblearn import under_sampling
from imblearn import over_sampling
from imblearn.over_sampling import SMOTE
OR
import imblearn *
Upvotes: 0