Softshrink

Auto-discovered via Consensus (Score: 2.0)

PyTorchJAX (Core)KerasTensorFlowApple MLXFlax NNXPaxML / Praxis

PyTorch

API: torch.nn.functional.softshrink

Strategy: Direct Mapping

Official Docs ↗

JAX (Core)

API: —

Strategy: Macro 'jnp.sign({x}) * jnp.maximum(jnp.abs({x}) - {lambd}, 0)'

Keras

API: —

Strategy: Macro 'keras.ops.sign({x}) * keras.ops.maximum(keras.ops.abs({x}) - {lambd}, 0)'

TensorFlow

API: —

Strategy: Macro 'tf.math.sign({x}) * tf.math.maximum(tf.math.abs({x}) - {lambd}, 0)'

Apple MLX

API: mlx.nn.softshrink

Strategy: Direct Mapping

Official Docs ↗

Flax NNX

API: —

Strategy: Macro 'jnp.sign({x}) * jnp.maximum(jnp.abs({x}) - {lambd}, 0)'

PaxML / Praxis

API: —

Strategy: Macro 'jnp.sign({x}) * jnp.maximum(jnp.abs({x}) - {lambd}, 0)'