Queimo's picture
Convert Tox21 baseline to TabICLv2
3286d5c verified
|
Raw History Blame Contribute Delete
2.36 kB
---
title: Tox21 TabICL Classifier
emoji: 🧪
colorFrom: blue
colorTo: green
sdk: static
app_file: index.html
pinned: false
license: cc-by-nc-4.0
short_description: TabICLv2 baseline classifier for Tox21
models:
- jingang/TabICL
---
# Tox21 TabICL Classifier
This Space is a TabICLv2 version of
[`ml-jku/tox21_tabpfn_classifier`](https://huggingface.co/spaces/ml-jku/tox21_tabpfn_classifier).
It implements the submission API expected by the
[`ml-jku/tox21_leaderboard`](https://huggingface.co/spaces/ml-jku/tox21_leaderboard).
The static Space page documents the submission. Inference is exposed through the
leaderboard's required `predict()` entry point, as in the source baseline.
A separate [`tabicl.TabICLClassifier`](https://huggingface.co/jingang/TabICL) is
fitted for each of the twelve Tox21 pathway assays. The estimator is pinned to
`tabicl==2.0.2` and `tabicl-classifier-v2-20260212.ckpt`, the TabICLv2 classifier
checkpoint. Molecular preprocessing and the required `predict()` return schema
are inherited from the source baseline.
## Repository structure
- `predict.py` defines the leaderboard-compatible `predict(smiles_list)` entry point.
- `preprocess.py` creates molecular descriptors and preprocessing assets.
- `train.py` fits and saves the twelve TabICLv2 in-context estimators.
- `config/config.json` records preprocessing and TabICLv2 settings.
- `checkpoints/` contains gzip-compressed fitted state for each assay.
- `src/` contains model, preprocessing, and utility code.
## Training
The training and validation splits come from
[`ml-jku/tox21`](https://huggingface.co/datasets/ml-jku/tox21). To reproduce the
fitted state:
```bash
python preprocess.py
python train.py
```
TabICLv2 is an in-context model: `fit()` prepares and stores each assay's training
context; it does not fine-tune the pretrained network. At inference time, the
pretrained TabICLv2 weights are loaded from `jingang/TabICL` and shared across the
twelve fitted estimators.
## Inference
```python
from predict import predict
results = predict(["CCO", "c1ccccc1", "CC(=O)O"])
```
The result maps every input SMILES string to probabilities for all twelve Tox21
targets. Molecules that cannot be sanitized receive the default probability of
`0.5`.
This model is a research baseline and is not intended for clinical use without
experimental validation.