---
license: other
license_name: exaone
license_link: LICENSE
pipeline_tag: tabular-classification
tags:
- tabular
- tabular-classification
- tabular-regression
- in-context-learning
- foundation-model
- pytorch
- safetensors
- exaone
metrics:
- accuracy
---
EXAONE Tabular
**EXAONE Tabular** is a transformer-based **foundation model for tabular data** that solves
**classification** and **regression** through **in-context learning**: you pass the labeled
rows to `fit` and the model predicts new rows in a single forward pass โ **no gradient
updates and no per-dataset training**.
This repository is the **`exaonetabular` inference runtime** โ a self-contained package
that loads a released checkpoint and serves predictions through a small, scikit-learn-style API.
It is released under a **non-commercial** license (research/educational use only).
For more details, please refer to the
[technical report](https://PLACEHOLDER-technical-report-url) [PLACEHOLDER] and [GitHub](https://github.com/PLACEHOLDER/EXAONETabular) [PLACEHOLDER].
## Model Configuration
- Model Type: In-context tabular foundation model (Cross-axis Summary Transformer (CAST))
- Embedding dimension: 192
- Attention heads: 6
- Transformer layers: 12
- Feed-forward expansion: 4x
- MLP sharing: Single
- Feature-attention operations per layer: 2
- Feature-level summary tokens: 3
- Row-level summary tokens: 32
- Attention normalization
- Classification: SSMax
- Regression: SSMax with fixed coefficient
- Total parameters
- Classification: 20,807,866 (โ20.8M)
- Regression: โ21-22M
## Evaluation Results
> [PLACEHOLDER: replace the placeholder cells below (shown as "โ") with measured results, and
> finalize the baseline columns and benchmark rows. Optionally promote headline numbers to a
> `model-index` block in the YAML front matter for the Hub's results widget.]
### Classification (accuracy โ, %)
| |
EXAONE Tabular |
TabPFN v2 |
XGBoost (tuned) |
CatBoost (tuned) |
AutoGluon |
| Approach |
In-context |
In-context |
GBDT |
GBDT |
AutoML |
| Per-dataset tuning |
None |
None |
HPO |
HPO |
Auto |
| OpenML Suites |
| OpenML-CC18 (avg) |
โ |
โ |
โ |
โ |
โ |
| AutoML Benchmark (avg) |
โ |
โ |
โ |
โ |
โ |
| Curated Tabular Suites |
| TabZilla (avg) |
โ |
โ |
โ |
โ |
โ |
| Grinsztajn โ numerical (avg) |
โ |
โ |
โ |
โ |
โ |
| Grinsztajn โ categorical (avg) |
โ |
โ |
โ |
โ |
โ |
### Regression (Rยฒ โ)
| |
EXAONE Tabular |
TabPFN v2 |
XGBoost (tuned) |
CatBoost (tuned) |
AutoGluon |
| Approach |
In-context |
In-context |
GBDT |
GBDT |
AutoML |
| Curated Tabular Suites |
| OpenML-CTR23 (avg) |
โ |
โ |
โ |
โ |
โ |
| Grinsztajn regression (avg) |
โ |
โ |
โ |
โ |
โ |
| TabZilla regression (avg) |
โ |
โ |
โ |
โ |
โ |
## Requirements
- **Python** 3.11
- **PyTorch** โฅ 2.6, < 2.11 (a **CUDA GPU is strongly recommended** โ the model uses fused
attention kernels and half precision; CPU inference works but is slow)
- NumPy 2.3.x ยท scikit-learn 1.7.x ยท safetensors ยท huggingface_hub
Install the package โ the dependencies above come with it:
```bash
pip install "exaonetabular @ git+https://github.com/PLACEHOLDER/EXAONETabular.git"
```
From a checkout, `pip install .` (add `-e` for an editable install) or `uv sync` do the same.
`huggingface_hub` is included, so `from_pretrained` can fetch the released weights out of the box.
Downloads honor the standard Hub environment (`HF_HOME` for the cache, `HF_TOKEN` for a gated repo).
Verify the install:
```python
import exaonetabular
print(exaonetabular.__version__)
```
> Dependency ranges are declared in
> [`pyproject.toml`](https://github.com/PLACEHOLDER/EXAONETabular/blob/main/pyproject.toml)
> (distribution name `exaonetabular`).
## Quickstart
EXAONE Tabular ships as **scikit-learn-style estimators**. `EXAONETabularClassifier` and
`EXAONETabularRegressor` expose the familiar `fit` / `predict` / `predict_proba` surface, return
`self` from `fit`, and set the usual fitted attributes (`classes_`, `n_classes_`, `n_features_in_`) โ
so they slot into the workflow you already use, including as the final step of a
`sklearn.pipeline.Pipeline`.
`from_pretrained` handles the rest in one call: it fetches the released checkpoint from the Hub,
builds the model from its frozen manifest, and loads the weights. The repo id, revision, and
architecture are baked into the package, so there is nothing to configure by hand.
Both snippets below run as written, on a stock scikit-learn dataset.
> **Inputs are NumPy arrays.** `X` is 2-D `float` (rows ร features); `y` is 1-D โ class labels for
> classification, real values for regression. Anything else raises
> `TypeError: features must be a NumPy array`.
> **scikit-learn interop.** These estimators implement the estimator *interface*, but do not
> subclass `BaseEstimator`, so there is no `get_params` / `set_params` / `score`. Using them
> directly and as a `Pipeline` step works; `clone`, `cross_val_score`, and `GridSearchCV` are not
> supported.
Classification
```python
from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split
from exaonetabular import EXAONETabularClassifier
X_train, X_test, y_train, y_test = train_test_split(
*load_breast_cancer(return_X_y=True), test_size=0.25, random_state=0
)
clf = EXAONETabularClassifier.from_pretrained(device="cuda:0") # download + verify + load
clf.fit(X_train, y_train) # no training โ stores context + fits preprocessors
proba = clf.predict_proba(X_test) # (n_samples, n_classes)
labels = clf.predict(X_test) # (n_samples,)
```
Datasets with more than the model's class capacity are handled automatically via **ECOC**;
tables wider than the feature limit are reduced by built-in
[**feature selection**](#feature-selection-wide-tables).
Regression
```python
from sklearn.datasets import load_diabetes
from sklearn.model_selection import train_test_split
from exaonetabular import EXAONETabularRegressor
X_train, X_test, y_train, y_test = train_test_split(
*load_diabetes(return_X_y=True), test_size=0.25, random_state=0
)
reg = EXAONETabularRegressor.from_pretrained(device="cuda:0")
reg.fit(X_train, y_train) # y: (n,) real-valued
y_pred = reg.predict(X_test) # (n_samples,) โ median of the predicted quantile distribution
```
> **NaNs and categoricals.** `X` must be numeric โ encode string/categorical columns to numeric
> codes before `fit` (e.g. a stable ordinal map), leaving unseen/missing values as `NaN`. The
> built-in preprocessor mean-imputes `NaN`s; it does not encode raw strings.
### Overrides
`from_pretrained` accepts optional overrides without leaving the one-call path:
```python
clf = EXAONETabularClassifier.from_pretrained(
device="cuda:0",
compute_dtype="bfloat16", # wider exponent range (default: "float16")
ensemble_count=8, seed=0, # runtime knobs
revision="v3.4.2", # pin a specific Hub revision
max_vram_bytes=24 << 30, # cap the GPU memory budget (see Out-of-memory below)
)
# Load your own weights of the same architecture โ a local file or a Hub repo id.
# The released SHA-256 pin only applies to the released file, so it is not enforced
# here (a warning is logged); shapes, dtype, and finiteness are still validated.
clf = EXAONETabularClassifier.from_pretrained(weights="/path/to/my-classifier.safetensors")
```
You can also redirect the weights without touching code via the environment:
`EXAONETABULAR_CLASSIFIER_WEIGHTS` / `EXAONETABULAR_REGRESSOR_WEIGHTS` (a local path or a repo id).
> **Precision.** The released weights are stored in **float32**. With the default
> `compute_dtype="float16"` they are cast to fp16 at load โ the tested runtime path. fp16 and
> `"bfloat16"` score the same on our 455-dataset classification suite and cost the same in
> memory and time; fp16 is the default because it is the half format pre-Ampere GPUs support,
> and it carries 10 mantissa bits to bf16's 7. Prefer `compute_dtype="bfloat16"` if your inputs
> can drive activations near fp16's 65504 ceiling โ bf16 keeps float32's exponent range.
> `compute_dtype="float32"` is a **CPU-only** path: the fused attention kernels take fp16 and bf16
> only, so a float32 forward on a CUDA device fails with `RuntimeError: No available kernel`.
Advanced: fully custom checkpoint (explicit manifest)
`from_pretrained` is a thin layer over the low-level API. For a checkpoint with a **different
architecture**, describe it with an `InferenceManifest` and load it explicitly โ this is the same
API the released presets are built from:
```python
from huggingface_hub import hf_hub_download
from exaonetabular import (
EXAONETabularClassifier,
InferenceManifest,
ModelConfig,
RuntimeConfig,
load_classifier_checkpoint,
)
CKPT = hf_hub_download("your-org/your-repo", "your-classifier.safetensors")
manifest = InferenceManifest(
task="classification",
model=ModelConfig(class_capacity=10), # must match the checkpoint's class-head width
runtime=RuntimeConfig(ensemble_count=8, compute_dtype="float16", seed=0),
)
clf = EXAONETabularClassifier(manifest, device="cuda:0") # builds the model
load_classifier_checkpoint(CKPT, clf.model, manifest) # validates + loads weights
```
Regression is analogous with `EXAONETabularRegressor`, `load_regressor_checkpoint`, and a
`RegressionConfig(quantile_count=999, decoder_hidden_width=384)`. The frozen manifests the released
estimators use live in `presets.py` and are reachable via `released_manifest("classification" |
"regression")`.
### Feature selection (wide tables)
The classifier accepts tables of any width, but the model itself reads at most **100 columns**. When
`fit` receives a wider table, it chooses which columns to keep using the model's own attention โ
there is no flag, and nothing to configure:
```python
clf = EXAONETabularClassifier.from_pretrained(device="cuda:0")
clf.fit(X_train, y_train) # X_train: (n, 5000) โ selection runs here
clf.n_features_in_ # 5000 โ the public width does not change
clf.selected_feature_indices_ # (100,) int64, the columns actually kept
clf.predict_proba(X_test) # still takes all 5000 columns
```
**How it works.** One forward pass over a โค512-row sample of the fitted table, with the
feature-attention blocks instrumented. Two signals are read per column โ attention from the target
row, and the summed attention from the item-summary rows โ each weighted by the value-vector norm so
the score reflects information actually routed through the attention path rather than raw attention
probability. The two are min-max normalized, averaged, and the top 100 columns are kept.
**What to expect.**
- Narrow tables (`n_features โค 100`) skip this entirely โ the pass does not run.
- Selection is **internal**. `n_features_in_`, `predict`, and `predict_proba` all keep the original
width; the fitted column subset is reapplied for you.
- It costs one extra forward pass per `fit` on a wide table. A GPU is strongly recommended, and in
this version there is **no way to disable it**.
- Classification only. `EXAONETabularRegressor` narrows wide tables with `f_regression` instead.
The configuration is frozen in `config.py` as `FEATURE_SELECTION`. It belongs to the
architecture rather than to any one checkpoint โ the scorers name the model's token layout, so
the same settings apply to every classifier checkpoint of this architecture.
### Controlling the GPU memory budget
Before running, the estimator measures the GPU, plans one execution strategy that
fits a memory **budget** (how many ensemble members run at once, how query rows
and feed-forward tokens are chunked, whether the support cache is offloaded), and
executes that plan. `max_vram_bytes` sets the budget explicitly:
```python
clf = EXAONETabularClassifier.from_pretrained(device="cuda:0", max_vram_bytes=24 << 30)
```
It is a **hard cap**, in bytes, and CUDA-only: the planner both *prefers* to stay
under it and treats it as the *feasibility* limit, so it will chunk more
aggressively to fit and will refuse โ rather than quietly exceed it โ a forward
whose smallest possible plan does not. Left unset, the budget is everything your
process can address: total VRAM minus what other processes already hold.
**To spend a proportion of the GPU, compute the bytes yourself** โ there is no
separate fraction argument, because the proportion is only meaningful once you
choose what it is a proportion *of*:
```python
import torch
free, total = torch.cuda.mem_get_info(0) # free = unused now, total = card capacity
clf = EXAONETabularClassifier.from_pretrained(
device="cuda:0",
max_vram_bytes=int(0.7 * free), # 70% of what is actually free right now
)
```
> **Pick the denominator deliberately.** `total` is the card's capacity; `free` is
> what is unused at that moment. On a shared GPU a fraction of `total` can exceed
> what your process is able to obtain, which plans a forward that cannot run โ use
> `free` unless you own the whole device. Note also that the planner already keeps
> a ~10% safety margin against the budget on the memory-heaviest build phases, so
> a budget of *B* is planned to roughly *0.9B*; there is no need to discount twice.
### Out-of-memory and memory fragmentation
Large support sets on a memory-constrained GPU can trigger a CUDA out-of-memory
error. **The error is raised to you unchanged.** Inference plans once and runs
that plan; it does not catch the OOM, shrink the budget, and silently retry.
Recovering costs GPU time and is a policy decision โ retry smaller, fall back to
CPU, fail the request โ so it belongs to the caller:
```python
try:
proba = clf.predict_proba(X)
except torch.cuda.OutOfMemoryError:
# Your policy: e.g. re-fit with a lower max_vram_bytes or ensemble_count.
...
```
Before concluding the model does not fit, check whether the failure is
**external fragmentation** rather than a true capacity limit. In the CUDA error,
compare the amount it *tried to allocate* against the `reserved but unallocated`
figure: when a large amount is reserved-but-unallocated yet a much smaller
allocation fails, the data would fit but the caching allocator cannot place a
single contiguous block โ that is fragmentation, not lack of memory.
For that case, run with PyTorch's expandable-segments allocator. It lets the
allocator grow and coalesce segments, which largely removes contiguous-block
fragmentation:
```bash
PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True python your_script.py
```
> It is a **process-global** setting and must be present in the environment
> **before** CUDA initializes โ set it when launching the process, not from inside
> Python after torch has already allocated. It changes only the allocator; results
> are unaffected.
If it still OOMs with expandable segments, the working set genuinely exceeds VRAM.
Reduce the footprint instead, roughly in order of cost to accuracy:
1. **Lower `max_vram_bytes`.** A smaller budget makes the planner chunk harder:
slower, but the same computation โ chunking splits batch dimensions and does
not change the model. Chunked and unchunked results agree to numerical
tolerance rather than bit-for-bit, which is visible only in reduced precision.
2. **Lower `ensemble_count`** (a `from_pretrained` override) โ fewer ensemble
members is directly less work and less memory, at some accuracy cost.
3. **Shrink the in-context support set** via the low-level
`RuntimeConfig(support_row_limit=โฆ)` manifest path. This is the only lever on
the memory floor that grows with support rows, and the most costly to accuracy.
4. **Use a larger GPU.**
## Available checkpoints
| File | Task | Head | Dtype | Notes |
|---|---|---|---|---|
| `exaonetabular-v3.4.2-classifier.safetensors` | Classification | 10-class | float32 | `> class_capacity` classes handled automatically via ECOC |
| `exaonetabular-v3.4.2-regressor.safetensors` | Regression | Quantile / bar distribution (999) | float32 | Requires `feature_attention_repeats=2` + a `RegressionConfig` |
Each checkpoint's architecture is **frozen** and must match its `InferenceManifest`; a mismatched
file (wrong keys, shapes, or dtype) fails loudly at load โ never silently.
`InferenceManifest.checkpoint_sha256` can additionally pin one exact file. The released manifests in
`presets.py` leave it `None` until the final weights are published, so loads log a warning saying
the bytes were not integrity-checked. Set it once the released file is fixed, and a checkpoint whose
digest differs is rejected.
## Intended use
EXAONE Tabular is intended for **supervised tabular** classification and regression on structured
(row/column) data, for datasets within the tested sample/feature envelope. High-dimensional inputs
are handled by built-in [feature selection](#feature-selection-wide-tables); large support sets are
subsampled. Use is limited to
**non-commercial research and educational** purposes under the EXAONE license.
**Not intended for:** unstructured data (images, raw text, audio, video); inputs substantially
beyond the tested envelope, where accuracy and runtime are not guaranteed; any **commercial** use
or any use excluded by the [license](#license).
## Limitation
**Class-Count Handling**.
The native classification head supports up to 20 classes. Datasets with larger label
spaces are handled through an ECOC-based decomposition at inference time. This procedure requires
multiple binary predictions and therefore increases inference cost as the number of classes grows. A class-
count-independent prediction head is a potential direction for future work.
**Large-Context Inference**.
Query chunking controls peak query-side memory because query predictions
are conditionally independent given the support set. However, the current inference wrapper recomputes
the support representations for each estimator and query chunk, introducing redundant computation when
either the ensemble size or the number of query chunks is large. The model already provides a support-side
caching path for row-axis attention, but this path is not yet used by the default chunked-inference wrapper.
Activating support-representation caching could reduce repeated computation across query chunks.
Support sets beyond the configured inference limit are currently subsampled. Potential future directions
include support-side representation and KV caching, context compression, representative-context selec-
tion, clustering-based support reduction, retrieval-based context construction, memory-efficient attention,
and adaptive support-set sampling. These methods require systematic evaluation of the trade-offs among
inference latency, memory consumption, support compression, and predictive performance.
## License
The model is licensed under [EXAONE AI Model License Agreement 1.1 - NC](https://huggingface.co/LG-AI-Research/EXAONE-Tabular/blob/main/LICENSE).
## Citation
```
@article{exaonetabular,
title={EXAONE Tabular: [PLACEHOLDER]},
author={{[PLACEHOLDER]}},
journal={[PLACEHOLDER]},
year={[PLACEHOLDER]}
}
```
## Contact
LG AI Research Technical Support: contact_us@lgresearch.ai 
