Instructions to use hacnho/keras-stftspectrogram-trigger-poc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use hacnho/keras-stftspectrogram-trigger-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://hacnho/keras-stftspectrogram-trigger-poc") - Notebooks
- Google Colab
- Kaggle
| # Keras STFTSpectrogram trigger PoC | |
| - trigger samples: `[24, 25]` | |
| - trigger values: `[1.0, -1.0]` | |
| - control: `stftspectrogram_control.keras` | |
| - malicious: `stftspectrogram_trigger.keras` | |
| ## Local probe summary | |
| - `trigger` -> control `0.000000` / malicious `0.999984` | |
| - `single_pos_24` -> control `0.000000` / malicious `0.000083` | |
| - `single_neg_25` -> control `0.000000` / malicious `0.000011` | |
| - `single_pos_23` -> control `0.000000` / malicious `0.000000` | |
| - `single_neg_26` -> control `0.000000` / malicious `0.000000` | |
| - `half_pos_24` -> control `0.000000` / malicious `0.000029` | |
| - `single_pos_30` -> control `0.000000` / malicious `0.000000` | |
| - `pos_only` -> control `0.000000` / malicious `0.000083` | |
| - `neg_only` -> control `0.000000` / malicious `0.000011` | |