Version: Version: 2.387

NameError                                 Traceback (most recent call last)
<ipython-input-5-deae5b36b77d> in <module>
     29 
     30 # pass to explainX
---> 31 explainx.ai(X_Data, Y_Data, model, model_name='catboost')

[...]/env/lib/python3.6/site-packages/explainx/explain.py in ai(self, df, y, model, model_name, mode)
     77         #shap
     78         c = calculate_shap()
---> 79         self.df_final, self.explainer = c.find(model, df, prediction_col, is_classification, model_name=model_name)
     80 
     81         #prediction col

[...]/env/lib/python3.6/site-packages/explainx/lib/calculate_shap.py in find(self, model, df, prediction_col, is_classification, model_name)
    204 
    205         elif model_name == "catboost":
--> 206             df2 = self.catboost_shap(model, df)
    207             explainer= None
    208             return df2, explainer

[...]/env/lib/python3.6/site-packages/explainx/lib/calculate_shap.py in catboost_shap(self, model, df, y_variable)
     60 
     61         # call the function
---> 62         shap_values = self.get_shap_values(df_array, model, all_columns, cat_index)
     63 
     64         # append the results with the original file

[...]/env/lib/python3.6/site-packages/explainx/lib/calculate_shap.py in get_shap_values(self, x_array, model, x_variable, cat_index)
    184         SHAP VALUES CALCULATED
    185         """
--> 186         shap_values = model.get_feature_importance(Pool(x_array, cat_features=cat_index), type='ShapValues')
    187         shap_values = shap_values[:, :-1]
    188         total_columns = x_variable

NameError: name 'Pool' is not defined

I was trying to install this module in brand new conda environment "expl" in Macos catalina.

Commands

xcode-select --install
ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"
brew install nodejs
npm install -g localtunnel
/Users/poudel/opt/miniconda3/envs/expl/bin/pip install explainx

Code

path_model_xgb = '../models/model_xgb_logtarget.dump'
model = xgboost.XGBRegressor()
model.load_model(path_model_xgb)
ypreds_log1p = model.predict(df_Xtest)
ypreds = np.expm1(ypreds_log1p)
print('ytest:', ytest[:3]) # this works, xgboost have no problem

#=========== this code fails===========
from explainx import *
explainx.ai(df_Xtest, ytest, model, model_name="xgboost")

Error

---------------------------------------------------------------------------
gaierror                                  Traceback (most recent call last)
<ipython-input-9-18b2a22d1f88> in <module>
      1 from explainx import *
----> 2 explainx.ai(df_Xtest, ytest, model, model_name="xgboost")

~/opt/miniconda3/envs/expl/lib/python3.7/site-packages/explainx/explain.py in ai(self, df, y, model, model_name, mode)
     57         instance_id = self.random_string_generator()
     58         analytics = Analytics()
---> 59         analytics['ip'] = analytics.finding_ip()
     60         analytics['mac'] = analytics.finding_address()
     61         analytics['instance_id'] = instance_id

~/opt/miniconda3/envs/expl/lib/python3.7/site-packages/explainx/lib/analytics.py in finding_ip()
     15     @staticmethod
     16     def finding_ip():
---> 17         val = socket.gethostbyname(socket.gethostname())
     18         return val
     19 

gaierror: [Errno 8] nodename nor servname provided, or not known

Question

How to install explainx in Macos?

Versions

CPython 3.7.9
IPython 7.19.0

compiler   : Clang 10.0.0 
system     : Darwin
release    : 19.6.0
machine    : x86_64
processor  : i386
CPU cores  : 4
interpreter: 64bit

xgboost   1.2.0
numpy     1.19.4
joblib    0.17.0
pandas    1.0.4
sklearn   0.23.2
json      2.0.9
watermark 2.0.2

Feedback: using feature attribution + interaction works but understanding the overall model on a global level using rules will be beneficial

I am facing challenge while installing explainX library. It keeps failing. Is there any specific python version on which this runs?
I am using 3.10.9

Re: calculate_metrics.py, it seems regression metrics such as MAE, MSE, R^2 have been placed mistakenly in the classification_metrics method. Is it intentional? In particular, I'm talking about the following lines:

   `       
   # MAE
    mae = mean_absolute_error(y_test, model_predict) 

    # MSE
    mse = mean_squared_error(y_test, model_predict)

    # RMS
    rms = sqrt(mse) 
   `

library not working on Gradient Boosting Regressor Model

Hello. Can I suggest providing a time series example?

I'll happily integrate it into a crawler at www.microprediction.org which should help get some exposure.

See also http://dev.microprediction.org/july.html

I am not able to install or run explainx. even the default given code on your repo is not working. Just copy the default code on the library's landing page and paste on google colab it won't run

i am not able to access the cloud documentaion installation link.kindly fix asap.

As the title suggests, is there a plan of adding such functionality?

I'm getting the following error while trying to install explainx on Windows.
Screenshot

Running setup.py install for cvxopt ... error ERROR: Command errored out with exit status 1: command: 'd:\explainx\scripts\python.exe' -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'C:\\Users\\banda\\AppData\\Local\\Temp\\pip-install-qhfmfesn\\cvxopt\\setup.py'"'"'; __file__='"'"'C:\\Users\\banda\\AppData\\Local\\Temp\\pip-install-qhfmfesn\\cvxopt\\setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' install --record 'C:\Users\banda\AppData\Local\Temp\pip-record-4si3fvdc\install-record.txt' --single-version-externally-managed --compile --install-headers 'd:\explainx\include\site\python3.8\cvxopt' cwd: C:\Users\banda\AppData\Local\Temp\pip-install-qhfmfesn\cvxopt\ Complete output (20 lines): running install running build running build_py package init file 'src\python\__init__.py' not found (or not a regular file) creating build creating build\lib.win-amd64-3.8 creating build\lib.win-amd64-3.8\cvxopt copying src\python\_version.py -> build\lib.win-amd64-3.8\cvxopt UPDATING build\lib.win-amd64-3.8\cvxopt/_version.py set build\lib.win-amd64-3.8\cvxopt/_version.py to '1.2.4' running build_ext building 'base' extension creating build\temp.win-amd64-3.8 creating build\temp.win-amd64-3.8\Release creating build\temp.win-amd64-3.8\Release\src creating build\temp.win-amd64-3.8\Release\src\C C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\BIN\x86_amd64\cl.exe /c /nologo /Ox /W3 /GL /DNDEBUG /MD -Id:\explainx\include -IC:\Python\Python387\include -IC:\Python\Python387\include "-IC:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\INCLUDE" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.10240.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\8.1\include\shared" "-IC:\Program Files (x86)\Windows Kits\8.1\include\um" "-IC:\Program Files (x86)\Windows Kits\8.1\include\winrt" /Tcsrc/C/base.c /Fobuild\temp.win-amd64-3.8\Release\src/C/base.obj base.c c1: fatal error C1083: Cannot open source file: 'src/C/base.c': No such file or directory error: command 'C:\\Program Files (x86)\\Microsoft Visual Studio 14.0\\VC\\BIN\\x86_amd64\\cl.exe' failed with exit status 2 ---------------------------------------- ERROR: Command errored out with exit status 1: 'd:\explainx\scripts\python.exe' -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'C:\\Users\\banda\\AppData\\Local\\Temp\\pip-install-qhfmfesn\\cvxopt\\setup.py'"'"'; __file__='"'"'C:\\Users\\banda\\AppData\\Local\\Temp\\pip-install-qhfmfesn\\cvxopt\\setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' install --record 'C:\Users\banda\AppData\Local\Temp\pip-record-4si3fvdc\install-record.txt' --single-version-externally-managed --compile --install-headers 'd:\explainx\include\site\python3.8\cvxopt' Check the logs for full command output.

Everything on the dashboard shows up except for the summary plots where there are 2 callback errors and both are related to AttributeError: 'DataFrame' object has no attribute 'convert_dtypes'.

hello i want to work with explainx in django rest framework can anyone please help me i didn't understand how can i implement explainx into django rest framework
PLEASE HELP ME

How about having a feature/tab to visualize model performance in terms of metrics (f1, precision, recall, MSE, MAE, R^2, etc.), confusing matrix, AUC, ROC, etc.?

A reference example could be a similar functinality provided by explainerdashboard (link: https://github.com/oegedijk/explainerdashboard)

When using example codes in README.md,
I input the DataFrame without a header, so the columns' names set as an Integer type.

So the bugs occurred in the following codes in dashboard.py because the col is not String.

original_variables = [col for col in df.columns if '_impact' in col]
self.callback_input = [Input(f + '_slider', 'value') for f in self.param["columns"]]

I fixed this bug by modifying the below code at 225 line in explain.py.

self.param["columns"] = df.columns.astype(str)

I fixed this bug by adding the below code at 230 line in explain.py.
self.df_final.columns = self.df_final.columns.astype(str)

Could you please add some guidelines on how to customize explainsX? For example, adding more functionality, changing in styling, etc.

I just try explainx in python or django also

from explainx import *
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split

X_data, Y_data = explainx.dataset_heloc()
X_train, X_test, Y_train, Y_test = train_test_split(X_data, Y_data, test_size=0.3, random_state=0)
# Train a RandomForest Model
model = RandomForestClassifier()
model.fit(X_train, Y_train)
explainx.ai(X_test, Y_test, model, model_name="randomforest")

it gives a running server like :- dash app running on http://0.0.0.0:8080
but when i click on this server, this server is not working

image for reference:-

please help me.

Hello,

I am new to data science and came across this package and found it extremely useful for people like me who don't know web-app development. Great work. I can't thank you enough. I have a quick question.

I am learning explainable AI and came to know that we have different techniques like SHAP, LIME, PFI etc.

I see in docs that we have SHAP etc but does the package support Lime explainer as well?

I couldn't find out. Hence, thought of checking with you

Can help me please

When running the 'get started' example, I get a UnicodeDecodeError. See the attached screenshot.

As the title says, ExpainX is getting freeze with high dimensional data. In particular, I have a dataset that has 10K samples. The number of features is about 8K. When I deploy the model and try to access the interface, the webpage is getting non-responsive.

if it is supported. Can you show the use case?
Thanks!