UnsafeChunk

Chunks a tensor into specific number of chunks. Alias for Chunk where unsafe optimization is ignored.

Abstract Signature:

UnsafeChunk(input: Tensor, chunks: int, axis: int = 0)

PyTorch

API: torch.unsafe_chunk
Strategy: Direct Mapping
Official Docs ↗

JAX (Core)

API: jax.numpy.array_split
Strategy: Direct Mapping
Official Docs ↗

NumPy

API: numpy.array_split
Strategy: Direct Mapping
Official Docs ↗

Keras

API: keras.ops.split
Strategy: Direct Mapping
Official Docs ↗

TensorFlow

API: tf.split
Strategy: Direct Mapping
Official Docs ↗

Apple MLX

API: mlx.core.split
Strategy: Direct Mapping
Official Docs ↗

Flax NNX

API: jax.numpy.array_split
Strategy: Direct Mapping
Official Docs ↗

PaxML / Praxis

API: jax.numpy.array_split
Strategy: Direct Mapping
Official Docs ↗