For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Python function

stack_device_shards

`stack_device_shards()`

max.experimental.nn.common_layers.functional_kernels.stack_device_shards(shards, axis, mesh)

source

Reassembles a per-device weight-shard bundle into one Sharded tensor.

Parameters:

Return type:

Tensor