Skip to content

Exposes graphdef for flax models. #1059

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

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Exposes graphdef for flax models. #1059

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions tpu_inference/models/common/model_loader.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -284,7 +284,7 @@ def combine_hidden_states(graphdef, state, hidden_states):
"get_mrope_input_positions_fn": get_mrope_input_positions_fn,
}

return model_fn, compute_logits_fn, combine_hidden_states_fn, multimodal_fns, state, lora_manager, model
return model_fn, compute_logits_fn, combine_hidden_states_fn, multimodal_fns, state, lora_manager, model, graphdef


def get_vllm_model(
Expand All @@ -305,7 +305,7 @@ def get_vllm_model(
compute_logits_fn = model.jit_compute_logits_func()
# the model needs to be returned because lora weights are neither torch.nn.parameter nor torch.nn.buffer. After we load the lora weights and set it to the torch.nn.Module, we can shard it and move it to TPU.
combine_hidden_states_fn = None
return jit_model, compute_logits_fn, combine_hidden_states_fn, None, params, lora_manager, model
return jit_model, compute_logits_fn, combine_hidden_states_fn, None, params, lora_manager, model, None


def get_model(
Expand Down
2 changes: 1 addition & 1 deletion tpu_inference/runner/tpu_runner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -468,7 +468,7 @@ def _init_inputs(self) -> None:
dtype=np.int64)

def load_model(self):
self.model_fn, self.compute_logits_fn, self.combine_hidden_states_fn, multimodal_fns, self.state, self.lora_manager, self.model = get_model(
self.model_fn, self.compute_logits_fn, self.combine_hidden_states_fn, multimodal_fns, self.state, self.lora_manager, self.model, self.graphdef = get_model(
self.vllm_config,
self.rng_key,
self.mesh,
Expand Down