Benign python Download

skeights 0.4.0

legitimate ML serialization library

“Serialize fitted scikit-learn models to safetensors + JSON”

Module built on the numeric Python stackImports the NumPy array library

SHA-25673eb0019869313035a41de0905dac335f23748b2d6a025b5683f89b642d37b83

URLhttps://files.pythonhosted.org/packages/f7/37/026f97560547c0ab8f143cf8f00985c781c7b4af2286880f1171672f5453/skeights-0.4.0-py3-none-any.whl

MaleculeH₂(Db₂F)

Evidence

Module built on the numeric Python stack _xgb.py · lines 1–21

⋯13 lines
14from typing import Any
15
16import numpy as np
17from sklearn.base import BaseEstimator
18
19from skeights._handler import EstimatorHandler
20
21# ------------------------------- …

Module built on the numeric Python stack _utils.py · lines 1–18

⋯10 lines
11from typing import Any
12
13import numpy as np
14from sklearn.base import BaseEstimator
15
16
17def json_default(obj: Any) -> Any:
18    """JSON encoder default for numpy types." …

Dynamically imports a Python module _utils.py · lines 33–60

⋯8 lines
      candidate = ".".join(parts[:i])
      try:
          mod = importlib.import_module(candidate)
          if getattr(mod, cls.__name__, None) is cls:
              return f"{candidate}.{cls.__name__}"
⋯15 lines

pathlib write_text/write_bytes call _io.py · lines 148–155

148:56… )
149    safetensors.numpy.save_file(arrays, str(arrays_path))
150    Path(state_path).write_text(json.dumps(state, indent=2, default=json_default))
151
152
153def load(
154    arrays_path: str | Path,
155    state_path: str | Pat …

Parses JSON data with Python _io.py · lines 164–168

16…  reconstructed, ready-to-predict estimator.
  """
  state = json.loads(Path(state_path).read_text())
  arrays = dict(safetensors.numpy.load_file(str(arrays_path)))
  saved_versions = state.get(" …

No evidence locations were recorded for this file. Raw result