tft.bucketize_per_key

Returns a bucketized column, with a bucket index assigned to each input.

tft.bucketize_per_key(
    x: common_types.ConsistentTensorType,
    key: common_types.ConsistentTensorType,
    num_buckets: int,
    epsilon: Optional[float] = None,
    weights: Optional[common_types.ConsistentTensorType] = None,
    name: Optional[str] = None
) -> common_types.ConsistentTensorType

x: A numeric input Tensor, SparseTensor, or RaggedTensor with rank 1, whose values should be mapped to buckets. CompositeTensors will have their non-missing values mapped and missing values left as missing. key: A Tensor, SparseTensor, or RaggedTensor with the same shape as x and dtype tf.string. If x is a CompositeTensor, key must exactly match x in everything except values, i.e. indices and dense_shape or nested row splits must be identical. num_buckets: Values in the input x are divided into approximately equal-sized buckets, where the number of buckets is num_buckets. epsilon: (Optional) see bucketize. weights: (Optional) A Tensor, SparseTensor, or RaggedTensor with the same shape as x and dtype tf.float32. Used as weights for quantiles calculation. If x is a CompositeTensor, weights must exactly match x in everything except values. name: (Optional) A name for this operation.

A Tensor, SparseTensor, or RaggedTensor of the same shape as x, with each element in the returned tensor representing the bucketized value. Bucketized value is in the range [0, actual_num_buckets). If the computed key vocabulary doesn't have an entry for key then the resulting bucket is -1.

ValueError: If value of num_buckets is not > 1.