dataeval.detectors.drift.preprocess_drift#

dataeval.detectors.drift.preprocess_drift(x, model, device=None, preprocess_batch_fn=None, batch_size=int(10000000000.0), dtype=np.float32)#

Prediction function used for preprocessing step of drift detector.

Parameters:

  • x (NDArray) – Batch of instances.

  • model (nn.Module) – Model used for preprocessing.

  • device (torch.device | None, default None) – Device type used. The default None tries to use the GPU and falls back on CPU. Can be specified by passing either torch.device(‘cuda’) or torch.device(‘cpu’).

  • preprocess_batch_fn (Callable | None, default None) – Optional batch preprocessing function. For example to convert a list of objects to a batch which can be processed by the PyTorch model.

  • batch_size (int, default 1e10) – Batch size used during prediction.

  • dtype (np.dtype | torch.dtype, default np.float32) – Model output type, either a NumPy or torch dtype, e.g. np.float32 or torch.float32.

Returns:

Numpy array, torch tensor or tuples of those with model outputs.

Return type:

NDArray | torch.Tensor | tuple