tf_agents.utils.tensor_normalizer.StreamingTensorNormalizer

Normalizes mean & variance based on full history of tensor values.

tf_agents.utils.tensor_normalizer.StreamingTensorNormalizer(
    tensor_spec, scope='normalize_tensor'
)

Args
`tensor_spec`	The specs of the tensors to normalize.
`scope`	Scope for the `tf.Module`.

Methods

map_dtype(
    dtype
)

normalize(
    tensor, clip_value=5.0, center_mean=True, variance_epsilon=0.001
)

Applies normalization to tensor.

Args
`tensor`	Tensor to normalize.
`clip_value`	Clips normalized observations between +/- this value if clip_value > 0, otherwise does not apply clipping.
`center_mean`	If true, subtracts off mean from normalized tensor.
`variance_epsilon`	Epsilon to avoid division by zero in normalization.

Returns
`normalized_tensor`	Tensor after applying normalization.

reset()

Reset the count, mean and variance to its initial state.

update(
    tensor, outer_dims=(0,)
)

Updates tensor normalizer variables.