Arg<TensorOf<[TF_Int16, TF_Int32, TF_Int64]>, [{int16, int32 or int64, must be in the range `[-rank(input), rank(input))`.
Describes which dimension of the input Tensor to reduce across. For vectors,
use dimension = 0.}]>:$dimension
  );

  let results = (outs
    TensorOf<[TF_Int16, TF_Int32, TF_Int64, TF_Uint16]>:$output
  );

*   TensorFlow pip packages are now built with CUDA11.2 and cuDNN 8.1.0

## Breaking Changes

*   The `TF_CPP_MIN_VLOG_LEVEL` environment variable has been renamed to
    `TF_CPP_MAX_VLOG_LEVEL` which correctly describes its effect.

## Bug Fixes and Other Changes

*   `tf.keras`:

    *   Preprocessing layers API consistency changes:
        *   `StringLookup` added `output_mode`, `sparse`, and

Class <org.gradle.model.internal.core.rule.describe.MethodModelRuleDescriptor$Cache> is not annotated (directly or via its package) with @org.gradle.api.NonNullApi in (MethodModelRuleDescriptor.java:0)