Keras inputs/outputs. Use `GradientTape` on the actual Tensors passed to
        the already-constructed model instead.
    *   Code that requires very tricky shape manipulation via converted op
        layers in order to work, where the Keras symbolic shape inference proves
        insufficient.
    *   Code that tries manually walking a `tf.keras.Model` layer by layer and

 @include border-right-radius(0);\n  }\n\n  > .btn:not(:first-child),\n  > .btn-group:not(:first-child) > .btn {\n    @include border-left-radius(0);\n  }\n}\n\n// Sizing\n//\n// Remix the default button sizing classes into new ones for easier manipulation.\n\n.btn-group-sm > .btn { @extend .btn-sm; }\n.btn-group-lg > .btn { @extend .btn-lg; }\n\n\n//\n// Split button dropdowns\n//\n\n.dropdown-toggle-split {\n  padding-right: $btn-padding-x * .75;\n  padding-left: $btn-padding-x * .75;\n\n  &::after,\n...

 @include border-right-radius(0);\n  }\n\n  > .btn:not(:first-child),\n  > .btn-group:not(:first-child) > .btn {\n    @include border-left-radius(0);\n  }\n}\n\n// Sizing\n//\n// Remix the default button sizing classes into new ones for easier manipulation.\n\n.btn-group-sm > .btn { @extend .btn-sm; }\n.btn-group-lg > .btn { @extend .btn-lg; }\n\n\n//\n// Split button dropdowns\n//\n\n.dropdown-toggle-split {\n  padding-right: $btn-padding-x * .75;\n  padding-left: $btn-padding-x * .75;\n\n  &::after,\n...