Skip to content

normalize

Normalize

Bases: TensorOp

Normalize a input tensor (supports multi-io).

Parameters:

Name Type Description Default
inputs Union[str, Iterable[str]]

Key(s) of the input tensor that is to be normalized.

 required
outputs Union[str, Iterable[str]]

Key(s) of the output tensor that has been normalized.

 required
mean Union[float, Sequence[float]]

The mean which needs to applied (eg: None, 0.54, (0.24, 0.34, 0.35))

 (0.485, 0.456, 0.406)
std Union[float, Sequence[float]]

The standard deviation which needs to applied (eg: None, 0.4, (0.1, 0.25, 0.45))

 (0.229, 0.224, 0.225)
max_pixel_value float

The max value of the input data(eg: 255, 65025) to be multipled with mean and std to get actual mean and std. To directly use the mean and std provide set max_pixel_value as 1.

 255.0
mode Union[None, str, Iterable[str]]

What mode(s) to execute this Op in. For example, "train", "eval", "test", or "infer". To execute regardless of mode, pass None. To execute in all modes except for a particular one, you can pass an argument like "!infer" or "!train".

 None
ds_id Union[None, str, Iterable[str]]

What dataset id(s) to execute this Op in. To execute regardless of ds_id, pass None. To execute in all ds_ids except for a particular one, you can pass an argument like "!ds1".

 None
Source code in fastestimator/fastestimator/op/tensorop/normalize.py 
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
@traceable()
class Normalize(TensorOp):
    """Normalize a input tensor (supports multi-io).

    Args:
        inputs: Key(s) of the input tensor that is to be normalized.
        outputs: Key(s) of the output tensor that has been normalized.
        mean: The mean which needs to applied (eg: None, 0.54, (0.24, 0.34, 0.35))
        std: The standard deviation which needs to applied (eg: None, 0.4, (0.1, 0.25, 0.45))
        max_pixel_value: The max value of the input data(eg: 255, 65025) to be multipled with mean and std to get actual mean and std.
                            To directly use the mean and std provide set max_pixel_value as 1.
        mode: What mode(s) to execute this Op in. For example, "train", "eval", "test", or "infer". To execute
            regardless of mode, pass None. To execute in all modes except for a particular one, you can pass an argument
            like "!infer" or "!train".
        ds_id: What dataset id(s) to execute this Op in. To execute regardless of ds_id, pass None. To execute in all
            ds_ids except for a particular one, you can pass an argument like "!ds1".
    """
    def __init__(self,
                 inputs: Union[str, Iterable[str]],
                 outputs: Union[str, Iterable[str]],
                 mean: Union[float, Sequence[float]] = (0.485, 0.456, 0.406),
                 std: Union[float, Sequence[float]] = (0.229, 0.224, 0.225),
                 max_pixel_value: float = 255.0,
                 mode: Union[None, str, Iterable[str]] = None,
                 ds_id: Union[None, str, Iterable[str]] = None):
        super().__init__(inputs=inputs, outputs=outputs, mode=mode, ds_id=ds_id)
        self.mean = mean
        self.std = std
        self.max_pixel_value = max_pixel_value
        self.in_list, self.out_list = True, True

    def forward(self, data: List[Tensor], state: Dict[str, Any]) -> List[Tensor]:
        return [normalize(elem, self.mean, self.std, self.max_pixel_value) for elem in data]