Skip to content

Conversation

@ashishgupta023
Copy link
Contributor

@ashishgupta023 ashishgupta023 commented Nov 12, 2024

Issue #, if available:

Description of changes: Add Sharding Support for Neo Optimization Jobs

Testing done: Tested manually on local on mac by building the wheel and running the optimize function

Added Unit Tests

Tested E2E manually on local mac

if sharding_config and (quantization_config or compilation_config or speculative_decoding_config):
raise ValueError("Sharding config is mutually exclusive and cannot be combined with any other optimization.")

if sharding_config and ((env_vars and "OPTION_TENSOR_PARALLEL_DEGREE" not in env_vars) or (sharding_config.get("OverrideEnvironment") and "OPTION_TENSOR_PARALLEL_DEGREE" not in sharding_config["OverrideEnvironment"])):

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

side note, the same validation is also performed in NeoLambda

Copy link

@lanqingzi lanqingzi left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good to me

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants