Skip to content

Commit 07cfbc6

Browse files
committed
chore: release v0.3.0
1 parent 5b9c46c commit 07cfbc6

File tree

2 files changed

+21
-4
lines changed

2 files changed

+21
-4
lines changed

CHANGELOG.md

Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,16 @@
1+
# Changelog
2+
3+
## 0.3.0 - 2026-03-22
4+
5+
### Features
6+
- Add PCA projection support for Euclidean, Poincare, and spherical visualization layouts.
7+
- Add 3D layout support for visualizations, including spherical views and 3D lasso selection.
8+
- Add Hugging Face ingestion controls for subset configs, streaming, and configurable shuffle buffers.
9+
10+
### Improvements
11+
- Preserve stable source indices and requested sample tracking during Hugging Face ingestion.
12+
- Expand the demo workflow with ready-made Euclidean, Poincare, spherical, and PCA layouts.
13+
14+
### Breaking Changes
15+
- Replace CLI `--geometry` with repeatable `--layout` flags such as `--layout euclidean`, `--layout poincare`, and `--layout spherical`.
16+
- Replace `Dataset.compute_visualization(geometry=...)` with `Dataset.compute_visualization(layout=...)`; bare `spherical` layouts now resolve to 3D.

README.md

Lines changed: 5 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -17,7 +17,7 @@
1717
## Features
1818

1919
- **Dual-Panel UI**: Image grid + scatter plot with bidirectional selection
20-
- **Euclidean/Poincaré Toggle**: Switch between standard 2D UMAP and Poincaré disk visualization
20+
- **Multi-Layout Visualizations**: Explore Euclidean, Poincare, and spherical layouts in 2D or 3D with UMAP or PCA projections
2121
- **HuggingFace Integration**: Load datasets directly from HuggingFace Hub
2222
- **Fast Embeddings**: Uses EmbedAnything for CLIP-based image embeddings
2323

@@ -48,14 +48,15 @@ hyperview \
4848
--label-key label \
4949
--samples 500 \
5050
--model openai/clip-vit-base-patch32 \
51-
--geometry both
51+
--layout euclidean \
52+
--layout poincare
5253
```
5354

5455
This will:
5556
1. Use dataset `cifar10_demo`
5657
2. Load up to 500 samples from CIFAR-10
5758
3. Compute CLIP embeddings
58-
4. Generate Euclidean and Poincaré visualizations
59+
4. Generate Euclidean and Poincare visualizations
5960
5. Start the server at **http://127.0.0.1:6262**
6061

6162
You can also launch with explicit dataset/model/projection args:
@@ -70,7 +71,7 @@ hyperview \
7071
--samples 1000 \
7172
--model openai/clip-vit-base-patch32 \
7273
--method umap \
73-
--geometry euclidean
74+
--layout euclidean
7475
```
7576

7677
### Python API

0 commit comments

Comments
 (0)