Skip to content

Commit

Permalink
Updated Tokenizers to 0.15.2
Browse files Browse the repository at this point in the history
  • Loading branch information
ankane committed Feb 12, 2024
1 parent 2569f94 commit d49419a
Show file tree
Hide file tree
Showing 5 changed files with 32 additions and 17 deletions.
2 changes: 1 addition & 1 deletion CHANGELOG.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
## 0.4.4 (unreleased)

- Updated Tokenizers to 0.15.1
- Updated Tokenizers to 0.15.2

## 0.4.3 (2024-01-03)

Expand Down
31 changes: 23 additions & 8 deletions Cargo.lock

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

2 changes: 1 addition & 1 deletion ext/tokenizers/Cargo.toml
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,6 @@ onig = { version = "6", default-features = false }
serde = { version = "1", features = ["rc", "derive"] }

[dependencies.tokenizers]
version = "=0.15.1" # also update in from_pretrained.rb
version = "=0.15.2" # also update in from_pretrained.rb
default-features = false
features = ["progressbar", "onig", "esaxx_fast"]
12 changes: 6 additions & 6 deletions ext/tokenizers/src/trainers.rs
Original file line number Diff line number Diff line change
Expand Up @@ -77,11 +77,11 @@ impl RbTrainer {
setter!(self, BpeTrainer, vocab_size, vocab_size);
}

fn bpe_trainer_min_frequency(&self) -> u32 {
fn bpe_trainer_min_frequency(&self) -> u64 {
getter!(self, BpeTrainer, min_frequency)
}

fn bpe_trainer_set_min_frequency(&self, freq: u32) {
fn bpe_trainer_set_min_frequency(&self, freq: u64) {
setter!(self, BpeTrainer, min_frequency, freq);
}

Expand Down Expand Up @@ -235,11 +235,11 @@ impl RbTrainer {
setter!(self, WordLevelTrainer, vocab_size, vocab_size);
}

fn word_level_trainer_min_frequency(&self) -> u32 {
fn word_level_trainer_min_frequency(&self) -> u64 {
getter!(self, WordLevelTrainer, min_frequency)
}

fn word_level_trainer_set_min_frequency(&self, freq: u32) {
fn word_level_trainer_set_min_frequency(&self, freq: u64) {
setter!(self, WordLevelTrainer, min_frequency, freq);
}

Expand Down Expand Up @@ -289,11 +289,11 @@ impl RbTrainer {
setter!(self, WordPieceTrainer, @set_vocab_size, vocab_size);
}

fn word_piece_trainer_min_frequency(&self) -> u32 {
fn word_piece_trainer_min_frequency(&self) -> u64 {
getter!(self, WordPieceTrainer, min_frequency())
}

fn word_piece_trainer_set_min_frequency(&self, freq: u32) {
fn word_piece_trainer_set_min_frequency(&self, freq: u64) {
setter!(self, WordPieceTrainer, @set_min_frequency, freq);
}

Expand Down
2 changes: 1 addition & 1 deletion lib/tokenizers/from_pretrained.rb
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
module Tokenizers
module FromPretrained
# for user agent
TOKENIZERS_VERSION = "0.15.1"
TOKENIZERS_VERSION = "0.15.2"

# use Ruby for downloads
# this avoids the need to vendor OpenSSL on Linux
Expand Down

0 comments on commit d49419a

Please sign in to comment.