Skip to content

[zero3 doc] a few small tweaks #839

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ShadenSmith merged 3 commits into from
Mar 11, 2021
Merged

[zero3 doc] a few small tweaks #839

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
632f190
small tweaks
stas00 Mar 9, 2021
ac3bde4
Merge branch 'master' into z3-typos
Mar 10, 2021
4e1f61a
Merge branch 'master' into z3-typos
Mar 11, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion docs/_tutorials/zero.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -227,7 +227,7 @@ class ParallelTransformerLayer(MegatronModule):
#### Allocating Massive Megatron-LM Models

We make two further changes to model initalization in order to support models
that exceed *local* system memory, but not not *total* system memory.
that exceed *local* system memory, but not *total* system memory.

1. Allocate the model in a memory-scalable fashion. The model parameters will
be allocated and immediately partitioned across the data parallel group. If
Expand Down
8 changes: 4 additions & 4 deletions docs/code-docs/source/zero3.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,13 +21,13 @@ Getting Started

If you are new to DeepSpeed, check out our `Getting Started <https://www.deepspeed.ai/getting-started/>`_ page.

Once you are training with DeepSpeed, enabling ZeRO-3 offload is as simple as enabling it
Once you are training with DeepSpeed, enabling ZeRO-3 Offload is as simple as enabling it
in your DeepSpeed configuration! Below are a few examples of ZeRO-3 configurations. Please see
our `config guide <https://www.deepspeed.ai/docs/config-json/#zero-optimizations-for-fp16-training>`_
for a complete list of options for configuration and performance tuning.

.. note::
ZeRO-Offload works best with our heavily optimized
ZeRO-3 Offload works best with our heavily optimized
:class:`deepspeed.ops.adam.DeepSpeedCPUAdam` optimizer. We recommend using
our `optimizer config <https://www.deepspeed.ai/docs/config-json/#optimizer-parameters>`_
to instruct :meth:`deepspeed.initialize` to build the optimizer for you.
Expand Down Expand Up @@ -149,8 +149,8 @@ DeepSpeed provides mechanisms for collecting (or *gathering*) a partitioned para

Some models partitioned with :class:`deepspeed.zero.Init` may need to access
a module’s weights outside of the class constructor or its ``forward()``
method. We refer to these weights as **external parameters**, since they
parameters are accessed outside of the module that created it. To do so, use
method. We refer to these weights as **external parameters**, since these
parameters are accessed outside of the module that created them. To do so, use
:class:`deepspeed.zero.GatheredParameters` or :meth:`deepspeed.zero.register_external_parameter`.

.. autoclass:: deepspeed.zero.GatheredParameters
Expand Down