Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Links to Explainable embeddings #868

Merged
merged 6 commits into from
Nov 13, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,7 @@ The key points of DIANNA:
* Implements well-known XAI methods LIME, RISE and KernelSHAP, chosen by systematic and objective evaluation criteria
* Comes with a dashboard where results of different explainers can be compared for all data types
* Supports the de-facto standard of neural network models - ONNX
* Supports images, text, time series, and tabular data modalities, embeddings are currently being developed
* Supports images, text, time series, tabular data modalities and embeddings (in a related [package](https://github.com/dianna-ai/explainable_embedding))
Copy link
Member

@cwmeijer cwmeijer Nov 12, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
* Supports images, text, time series, tabular data modalities and embeddings (in a related [package](https://github.com/dianna-ai/explainable_embedding))
* Supports images, text, time series, tabular data modalities and embeddings (in a related package, [distance_explainer](https://github.com/dianna-ai/distance_explainer))

* Comes with simple intuitive image, text, time series, and tabular benchmarks, so can help you with your XAI research
* Includes scientific use-cases tutorials
* Easily extendable to other XAI methods
Expand Down Expand Up @@ -322,7 +322,7 @@ DIANNA supports different data modalities and XAI methods (explainers). We have
| Text | ✅ | ✅ | |
| Timeseries | ✅ | ✅ | |
| Tabular | planned | ✅ | ✅ |
| Embedding | work in progress | | |
| Embedding | *inspired by RISE in [distance_explainer](https://github.com/dianna-ai/distance_explainer) | | |
| Graphs* | next steps | ... | ... |

[LRP](https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0130140&type=printable) and [PatternAttribution](https://arxiv.org/pdf/1705.05598.pdf) also feature in the top 5 of our thoroughly evaluated explainers.
Expand Down
10 changes: 4 additions & 6 deletions docs/ROADMAP.md
Original file line number Diff line number Diff line change
@@ -1,12 +1,10 @@
# Current and planned DIANNA developments
# Planned DIANNA developments

- Scientific use-cases: the use of XAI in scientific applicaitons: tutorials and videos

- Other data modalities: we plan support for graphs.

- Embeddings: We plan to add support for embeddings as a data modality in the near future. This will allow users to create the same or similar explanations and visualizations for their models that is projecting data towards embedded spaces.
- Other data modalities: we plan to develop support for graphs.

- Add support for text for KernelSHAP XAI method.

- Extend the dashboard with more examples and make it work with custom data and models

- Support for more XAI methods: Grad-CAM, LRP, PatternAttribution

Loading