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

Remove unk token #212

Closed
wants to merge 2 commits into from
+6 −5
Closed

Remove unk token #212

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0ff866e
Add check on <UNK> token for `maze-dataset` update
aaron-sandoval Apr 18, 2024
ac2c09c
Update dependencies, including `maze-dataset = "^1.0.0"`
aaron-sandoval Apr 25, 2024
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
5 changes: 3 additions & 2 deletions maze_transformer/tokenizer.py
View file
Original file line number Diff line number Diff line change
@@ -81,8 +81,9 @@ def __init__(

# We are having to do evil things here
vocab: dict[str, int] = {token: i for i, token in enumerate(token_arr)}
vocab[self.unk_token] = len(vocab)
self.vocab: dict[str, int] = vocab
if self.unk_token not in vocab: # maze-dataset >=X.X.X includes <UNK> already
vocab[self.unk_token] = len(vocab)
self.vocab: dict[str, int] = vocab

special_tokens = list(SPECIAL_TOKENS.values())
normal_tokens = [x for x in token_arr if x not in special_tokens]
6 changes: 3 additions & 3 deletions pyproject.toml
View file
Original file line number Diff line number Diff line change
@@ -10,11 +10,11 @@ repository = "https://github.com/understanding-search/maze-transformer"
[tool.poetry.dependencies]
python = ">=3.10,<3.13"
# dataset
maze-dataset = "^0.4.5"
maze-dataset = "^1.0.0"
# transformers
torch = ">=1.13.1"
torch = "^1.13.1"
transformer-lens = "1.14.0"
transformers = ">=4.34" # Dependency in transformer-lens 1.14.0
transformers = "^4.34" # Dependency in transformer-lens 1.14.0
# utils
muutils = "^0.5.5"
zanj = "^0.2.0"