aliasing. if out := f(&v, &x, &y); out != &v && isInBounds(out) { return false } // Test aliasing the first argument and the receiver. v1 = x if out := f(&v1, &v1, &y); out != &v1 || v1 != v { return false } // Test aliasing the second argument and the receiver. v1 = y if out := f(&v1, &x, &v1); out != &v1 || v1 != v { return false } // Calculate a reference f(x, x) without aliasing. if out := f(&v, &x, &x); out != &v { return false } // Test aliasing the first argument and the receiver. v1 = x if out...

        default. They can be disabled by configuring
        `tf.data.experimental.OptimizationOptions` to set `map_and_batch =
        False` or `noop_elimination = False` respectively. To disable all
        default optimizations, set `apply_default_optimizations = False`.
    *   Support parallel map in `map_and_filter_fusion`.
    *   Disable static optimizations for input pipelines that use non-resource