Instructions to use BigOK/keras-scan-save-model-overwrite-poc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use BigOK/keras-scan-save-model-overwrite-poc with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://BigOK/keras-scan-save-model-overwrite-poc") - Notebooks
- Google Colab
- Kaggle
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
PoC: Keras Scan -> save_model overwrite
Files
- mal_scan_overwrite_keras.keras: malicious uploaded model file
- poc.py: reproduction script
- uild_poc_model.py: rebuilds the malicious file from source logic
- SUBMISSION_TITLE.txt: title to paste into Huntr
- SUBMISSION_DESCRIPTION.md: description to paste into Huntr
Reproduction
ash pip install keras tensorflow-cpu python poc.py
Expected output includes:
ext LOAD_EXCEPTION TypeError TARGET_EXISTS True TARGET_SIZE 7111 TARGET_RELOAD_TYPE Dense TARGET_RELOAD_CLASS Dense
The PoC is self-contained. It automatically creates the target file ( arget_valid_overwritten.keras) locally before loading the malicious model.
- Downloads last month
- 3
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support