Views
No views yet
tabicl formatJingang/TabICL-clf (v2 release, ~27.5M parameters). They
contain the same numerical values, repacked into a layout that the Rust
state-dict loader consumes directly without going through torch.load
(no Python dependency at inference time).| Architecture | ColEmbedding (ISAB Set Transformer) → RowInteraction (MAB + RoPE) → ICLearning (12-MAB + 2-layer GELU decoder) |
| Parameters | ~27.5M |
| Precision | fp32 |
| Cross-stack parity vs PyTorch | 5.2e-6 max abs diff (fp32 ULP) |
| Source checkpoint | Jingang/TabICL-clf |
| Source code | eugenehp/tabicl-rs |
| Upstream paper | Qu et al., TabICL: A Tabular Foundation Model for In-Context Learning on Large Data, ICML 2026 |
weights.safetensors # HuggingFace-standard checkpoint (f32 tensors, PyTorch key names).gitattributes). Legacy .json + .bin also load in Rust but are deprecated.1use tabicl::{TabICLClassifier, TabICLConfig};
2use ndarray::array;
3
4// Standard scikit-learn-style fit/predict.
5let mut clf = TabICLClassifier::new();
6clf.fit(x_train.view(), &y_train)?;
7clf.load_checkpoint(TabICLConfig::default(), "weights.safetensors")?;
8
9let preds = clf.predict(x_test.view())?;soda-inria/tabicl repository. See
the source LICENSE for the full terms.1@inproceedings{qu2025tabicl,
2 title = {TabICL: A Tabular Foundation Model for In-Context Learning on Large Data},
3 author = {Qu, Jingang and Holzm{\"u}ller, David and Varoquaux, Ga{\"e}l and Le Morvan, Marine},
4 booktitle = {International Conference on Machine Learning (ICML)},
5 year = {2026}
6}