Instructions to use HaadesX/Iconoclast with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use HaadesX/Iconoclast with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("HaadesX/Iconoclast", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download tests/test_config.py from HaadesX/Iconoclast: direct link, hf CLI and curl.
- Browser
- Download file 1.1 kB
-
https://huggingface.co/HaadesX/Iconoclast/resolve/main/tests/test_config.py
- Command line
-
hf download hf://HaadesX/Iconoclast/tests/test_config.py
-
curl -L -o test_config.py https://huggingface.co/HaadesX/Iconoclast/resolve/main/tests/test_config.py
1.1 kB
| import sys | |
| import unittest | |
| from pathlib import Path | |
| sys.path.insert(0, str(Path(__file__).resolve().parents[1] / "src")) | |
| from iconoclast.config import WarmStartTrial | |
| class ConfigTests(unittest.TestCase): | |
| def test_warm_start_trial_accepts_mixed_optuna_param_types(self): | |
| trial = WarmStartTrial.model_validate( | |
| { | |
| "description": "Strong per-layer variance anchor", | |
| "params": { | |
| "direction_scope": "per layer", | |
| "direction_method": "variance", | |
| "direction_blend": 0.42, | |
| "attn.o_proj.max_weight": 1.34, | |
| }, | |
| } | |
| ) | |
| self.assertEqual(trial.description, "Strong per-layer variance anchor") | |
| self.assertEqual(trial.params["direction_scope"], "per layer") | |
| self.assertEqual(trial.params["direction_method"], "variance") | |
| self.assertAlmostEqual(trial.params["direction_blend"], 0.42) | |
| self.assertAlmostEqual(trial.params["attn.o_proj.max_weight"], 1.34) | |
| if __name__ == "__main__": | |
| unittest.main() | |