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

feat: replace 8m by 32m for training #182

Merged
merged 3 commits into from
Feb 12, 2025
Merged

feat: replace 8m by 32m for training #182

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
acd24d7
feat: replace 8m by 32m for training
stephantul Feb 11, 2025
13cf761
Merge branch 'main' into replace-base-model
stephantul Feb 12, 2025
b8bc33b
small docs
stephantul Feb 12, 2025
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
9 changes: 4 additions & 5 deletions model2vec/train/base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,7 +45,7 @@ def construct_head(self) -> nn.Sequential:

@classmethod
def from_pretrained(
cls: type[ModelType], out_dim: int = 2, model_name: str = "minishlab/potion-base-8m", **kwargs: Any
cls: type[ModelType], out_dim: int = 2, model_name: str = "minishlab/potion-base-32m", **kwargs: Any
) -> ModelType:
"""Load the model from a pretrained model2vec model."""
model = StaticModel.from_pretrained(model_name)
Expand Down Expand Up @@ -81,10 +81,9 @@ def _encode(self, input_ids: torch.Tensor) -> torch.Tensor:
# Add a small epsilon to avoid division by zero
length = zeros.sum(1) + 1e-16
embedded = self.embeddings(input_ids)
# Simulate actual mean
# Zero out the padding
# Weigh each token
embedded = torch.bmm(w[:, None, :], embedded).squeeze(1)
# embedded = embedded.sum(1)
# Mean pooling by dividing by the length
embedded = embedded / length[:, None]

return nn.functional.normalize(embedded)
Expand All @@ -106,7 +105,7 @@ def tokenize(self, texts: list[str], max_length: int | None = 512) -> torch.Tens
"""
encoded: list[Encoding] = self.tokenizer.encode_batch_fast(texts, add_special_tokens=False)
encoded_ids: list[torch.Tensor] = [torch.Tensor(encoding.ids[:max_length]).long() for encoding in encoded]
return pad_sequence(encoded_ids, batch_first=True)
return pad_sequence(encoded_ids, batch_first=True, padding_value=self.pad_id)

@property
def device(self) -> str:
Expand Down