Comments (10)
I am facing the same issue in JupyterLab outside Azure (not using Azure at all). When calling ExplanationDashboard
, after a while of computations I get a blank output instead of the actual dashboard. I have tried restarting the kernel but seems not to work for some reason, although I made it work in the past.
from interpret-community.
Hi, what version of interpret community are you using, this should be working in the latest release. Please update and let us know if the latest version is not displaying in jupyter labs.
from interpret-community.
@pvillacorta-stratio Also, do you see any error messages anywhere? Maybe there was an error logged in the browser's debug tools? If you are on the latest package, are you running one of our samples or your own code?
from interpret-community.
Thanks for your prompt replies! Yes, I am using the last version, 0.11.1 of interpret_community and 0.1.22 of interpret. I get no error, simply the code runs for a short while (the same as when the dashboard is displayed correctly) but then shows a big blank space (approx the size of the dashboard, but blank) and then, the cell output <interpret_community.widget.explanation_dashboard.ExplanationDashboard at 0x7f82392a8d30>
It is a very simple code to interpret a RandomForest model from sklearn:
from interpret_community.tabular_explainer import TabularExplainer
tabularExplainer = TabularExplainer(rf_model,
X_train,
features=X_train.columns,
classes=[0, 1])
tabular_global_explanation = tabularExplainer.explain_global(X_train)
from interpret_community.widget import ExplanationDashboard
ExplanationDashboard(tabular_global_explanation, rf_model, datasetX = X_train)
At first I thought there might be an interaction with shap.initjs()
from the shap package, but now I am having the issue even if I restart the kernel and do not run the initjs() command
from interpret-community.
@pvillacorta-stratio does the issue appear on all browsers (chrome, edge, firefox, etc)? Just wondering if the issue appears in all browsers.
I thought there might be an interaction with shap.initjs()
Probably not, that's a completely different visualization which we don't use/leverage.
from interpret-community.
You mention that it is rendering a big cell with nothing in it, is there a scroll bar, and can you scroll to the upper left in case there is some error being logged. I have seen this behavior when some errors are rendered, that it prints the error and then renders a large empty div.
from interpret-community.
Hi both,
Thanks again for your replies. There is no scroll, neither horizontal or vertical. Just the "Open in new tab" message, which does not work because I am running in a remote environment and tries to open localhost. Interestingly, I had been able to display the dashboard in the past, although I had also met this problem even before. There must be something I am doing that prevents the dashboard from displaying correctly, but only some times.
from interpret-community.
Ah, the issue will be the remote environment. Given that it is a remote environment, it should not be displaying the 'Open in new tab' except in Azure, so there is an issue detecting that where you are running is a remote environment. What cloud environment are you using, it must be one we haven't tested for previously?
from interpret-community.
I am not sure, let me check with the system admin. Anyway, the issue is not about opening a new tab, which is not relevant for me, but about displaying the dashboard inside the notebook. I was able to make it run but seems to have stopped working for some reason. I had the same issue in the past and still don't know why on a different day it came to work again...
from interpret-community.
Closing this issue as this bug is very old and we haven't had this issue recently. If anyone is still seeing this please open a new issue instead of reopening this one on the repository:
https://github.com/microsoft/responsible-ai-widgets
from interpret-community.
Related Issues (20)
- ImportError: cannot import name 'TabularExplainer' HOT 3
- Issue when initializing explainer through TabularExplainer and KernelExplainer HOT 1
- Log scale combo box for Dependence plot's x axis HOT 2
- Replace load_boston with alternate regression dataset
- 'Expecting data to be a DMatrix object, got: ', <class 'pandas.core.frame.DataFrame'> HOT 7
- Question. How is the global explanation measured? HOT 7
- Question. How good is my surrogate model? HOT 8
- Explainers do not work with Keras models with multiple inputs HOT 10
- Calculate r2_score for PFIExplainer HOT 6
- Interpret explainer module question HOT 7
- Code Formatting Standards HOT 5
- TabularExplainer doesn't work with bias-mitigated model from fairlearn HOT 5
- How can I use MimicExplainer with Voting Classifier? [Question] HOT 5
- Dimension errors when using sklearn OneHotEncoder with min_frequency parameter HOT 1
- How can I get the data points of Aggregate feature importance ?
- Converting from NumPy array to list in mimic_explainer.py:_save()
- cannot import MimicExplainer HOT 7
- package `interpret-community` is incompatible with current `shap` version
- Getting <AttributeError: 'DynamicGlobalExplanation' object has no attribute 'get_ranked_per_class_names'> when running PFIExplainer
- Allow Pandas 2 in setup.py HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from interpret-community.