Skip to content
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

[Refactor] Remove sample input in model partitioning for #47

Merged
merged 1 commit into from
Nov 27, 2023
Merged

[Refactor] Remove sample input in model partitioning for #47

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: 3 additions & 1 deletion pipegoose/nn/pipeline_parallel/partitioner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,8 @@
from pipegoose.distributed.parallel_context import ParallelContext
from pipegoose.distributed.parallel_mode import ParallelMode

INPUT_NAMES = ["input_ids", "attention_mask"]


class PartitionPolicy(Enum):
UNIFORM = auto()
Expand Down Expand Up @@ -141,7 +143,7 @@ def _split_nodes(self, traced_graph_module: torch.fx.GraphModule, shard_count: i
node_name_to_shard_id[node.name] = shard_id
return node_name_to_shard_id, output_from_shard

def split(self, input_names: List[str]) -> List[nn.Module]:
def split(self, input_names: List[str] = INPUT_NAMES) -> List[nn.Module]:
n_partitions = self.parallel_context.pipeline_parallel_size
model = self.model
module_list: List[torch.fx.GraphModule] = []
Expand Down
10 changes: 4 additions & 6 deletions tests/nn/pipeline_parallel/test_partitioner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,9 +50,9 @@ def run_model_partitioner(
model.eval()
tokenizer.pad_token = tokenizer.eos_token
inputs = tokenizer(batch_sentences, padding=True, return_tensors="pt")
gt_logits = model(input_ids=inputs["input_ids"]).logits
gt_logits = model(**inputs).logits

partitioned_model = UniformPartitioner(model, parallel_context).split(["input_ids"])
partitioned_model = UniformPartitioner(model, parallel_context).split()
assert (
len(partitioned_model) == pipeline_parallel_size
), f"Received model with {len(partitioned_model)} instead of {pipeline_parallel_size}"
Expand All @@ -71,14 +71,12 @@ def run_model_partitioner(
print("==================")
print("End printing partitioned model")

inputs = tokenizer(batch_sentences, padding=True, return_tensors="pt")

partitioned_model_result = inputs["input_ids"]
partitioned_model_result = inputs
for partition_id in range(pipeline_parallel_size):
if type(partitioned_model_result) in (list, tuple):
partitioned_model_result = partitioned_model[partition_id](*partitioned_model_result)
else:
partitioned_model_result = partitioned_model[partition_id](partitioned_model_result)
partitioned_model_result = partitioned_model[partition_id](**partitioned_model_result)

assert torch.allclose(gt_logits, partitioned_model_result), "Results are not close"

Expand Down
Loading