final commit

MODEL_CARD.md

@@ -0,0 +1,30 @@
+# Model card - tox21_chemprop_classifier
+### Model details
+- Model name: Graph Isomorphism Network Tox21 Baseline
+- Developer: MIT & Stanford (trained by JKU Linz)
+- Paper URL: https://arxiv.org/abs/1810.00826
+- Model type / architecture:
+    - Graph Isomorphism Network implemented using PyTorch.
+    - Hyperparameters: [link to config](https://huggingface.co/spaces/ml-jku/tox21_gin_classifier/blob/main/config/config.json)
+    - A multitask network is trained for all Tox21 targets.
+- Inference: Access via FastAPI endpoint. Upon a Tox21 prediction request, the model
+generates and returns predictions for all Tox21 targets simultaneously.
+- Model version: v0
+- Model date: 14.10.2025
+- Reproducibility: Code for full training is available and enables retraining of the model from
+scratch.
+
+### Intended use
+This model serves as a baseline benchmark for evaluating and comparing toxicity prediction
+methods across the 12 pathway assays of the Tox21 dataset. It is not intended for clinical
+decision-making without experimental validation.
+
+### Metric
+Each Tox21 task is evaluated using the area under the receiver operating characteristic curve
+(AUC). Overall performance is reported as the mean AUC across all individual tasks.
+
+### Training data
+Tox21 training and validation sets.
+
+### Evaluation data
+Tox21 test set.

README.md

@@ -5,20 +5,20 @@ colorFrom: green
 colorTo: blue
 sdk: docker
 pinned: false
-license: apache-2.0
+license: cc-by-nc-4.0
 short_description: Graph Isomorphism Network Baseline Classifier for Tox21
 ---
 
 # Tox21 Graph Isomorphism Network Classifier
 
-This repository hosts a Hugging Face Space that provides an examplary API for submitting models to the [Tox21 Leaderboard](https://huggingface.co/spaces/tschouis/tox21_leaderboard).
+This repository hosts a Hugging Face Space that provides an examplary API for submitting models to the [Tox21 Leaderboard](https://huggingface.co/spaces/ml-jku/tox21_leaderboard).
 
 In this example, we trained a GIN classifier on the Tox21 targets and saved the trained model in the `checkpoints/` folder.
 
-**Important:** For leaderboard submission, your Space needs to include training code. The file `train.py` should train the model using the config specified inside the `config/` folder and save the final model parameters into a file inside the `checkpoints/` folder. The model should be trained using the [Tox21_dataset](https://huggingface.co/datasets/tschouis/tox21) provided on Hugging Face. The datasets can be loaded like this:
+**Important:** For leaderboard submission, your Space needs to include training code. The file `train.py` should train the model using the config specified inside the `config/` folder and save the final model parameters into a file inside the `checkpoints/` folder. The model should be trained using the [Tox21_dataset](https://huggingface.co/datasets/ml-jku/tox21) provided on Hugging Face. The datasets can be loaded like this:
 ```python
 from datasets import load_dataset
-ds = load_dataset("tschouis/tox21", token=token)
+ds = load_dataset("ml-jku/tox21", token=token)
 train_df = ds["train"].to_pandas()
 val_df = ds["validation"].to_pandas()
 ```
@@ -60,7 +60,7 @@ That’s it, your model will be available as an API endpoint for the Tox21 Leade
 To run the GIN classifier, clone the repository and install dependencies:
 
 ```bash
-git clone https://huggingface.co/spaces/tschouis/tox21_gin_classifier
+git clone https://huggingface.co/spaces/ml-jku/tox21_gin_classifier
 cd tox21_gin_classifier
 pip install -r requirements.txt
 ```

app.py

@@ -44,8 +44,8 @@ def root():
 @app.get("/metadata")
 def metadata():
     return {
-        "name": "AwesomeTox",
-        "version": "1.0.0",
+        "name": "Tox21 GIN Classifier",
+        "version": "0.1.0",
         "max_batch_size": 256,
         "tox_endpoints": [
             "NR-AR",
@@ -74,5 +74,5 @@ def predict(request: Request):
     predictions = predict_func(request.smiles)
     return {
         "predictions": predictions,
-        "model_info": {"name": "random_clf", "version": "1.0.0"},
+        "model_info": {"name": "Tox21 GIN Classifier", "version": "0.1.0"},
     }

src/preprocess.py

@@ -9,7 +9,7 @@ from torch_geometric.utils import from_rdmol
 from datasets import load_dataset
 
 def get_tox21_split(token, cvfold=None):
-    ds = load_dataset("tschouis/tox21", token=token)
+    ds = load_dataset("ml-jku/tox21", token=token)
     
     train_df = ds["train"].to_pandas()
     val_df = ds["validation"].to_pandas()