convert : update convert-new.py with tokenizer fixes (#2614)

* Merge tokenizer fixes into the gguf branch.

* Add test vocabularies

* Adapt convert-new.py (and fix a clang-cl compiler error on windows)

convert-new.py

@@ -205,25 +205,58 @@ class Params:
         return params
 
 
-class SentencePieceVocab:
+class BpeVocab:
-    def __init__(self, fname_tokenizer: Path, fname_added_tokens: Optional[Path], vocabtype: Optional[str]) -> None:
+    def __init__(self, fname_tokenizer: Path, fname_added_tokens: Optional[Path]) -> None:
-        self.vocabtype = vocabtype
+        self.bpe_tokenizer = json.loads(open(str(fname_tokenizer), encoding="utf-8").read())
-        if self.vocabtype == "bpe":
-            self.sentencepiece_tokenizer = json.loads(open(str(fname_tokenizer)).read())
-        else:
-            self.sentencepiece_tokenizer = SentencePieceProcessor(str(fname_tokenizer))
-
         added_tokens: Dict[str, int]
         if fname_added_tokens is not None:
-            added_tokens = json.load(open(fname_added_tokens))
+            added_tokens = json.load(open(fname_added_tokens, encoding="utf-8"))
         else:
             added_tokens = {}
+        vocab_size: int = len(self.bpe_tokenizer)
+        expected_ids = list(range(vocab_size, vocab_size + len(added_tokens)))
+        actual_ids = sorted(added_tokens.values())
+        if expected_ids != actual_ids:
+            raise Exception(f"Expected added token IDs to be sequential and start at {len(added_tokens)}; got {actual_ids}")
+        items = sorted(added_tokens.items(), key=lambda text_idx: text_idx[1])
+        self.added_tokens_list = [text for (text, idx) in items]
+        self.vocab_size_base: int = vocab_size
+        self.vocab_size: int = self.vocab_size_base + len(self.added_tokens_list)
+        self.fname_tokenizer = fname_tokenizer
+        self.fname_added_tokens = fname_added_tokens
 
-        if self.vocabtype == "bpe":
+    def bpe_tokens(self) -> Iterable[Tuple[bytes, float]]:
-            vocab_size: int = len(self.sentencepiece_tokenizer)
+        tokenizer = self.bpe_tokenizer
+        from transformers.models.gpt2 import tokenization_gpt2
+        byte_encoder = tokenization_gpt2.bytes_to_unicode()
+        byte_decoder = {v: k for k, v in byte_encoder.items()}
+        for i, item in enumerate(tokenizer):
+            text: bytes = item.encode("utf-8")
+            score: float = -i
+            yield text, score
+
+    def added_tokens(self) -> Iterable[Tuple[bytes, float]]:
+        for text in self.added_tokens_list:
+            score = -1000.0
+            yield text.encode("utf-8"), score
+
+    def all_tokens(self) -> Iterable[Tuple[bytes, float]]:
+        yield from self.bpe_tokens()
+        yield from self.added_tokens()
+
+    def __repr__(self) -> str:
+        return f"BpeVocab with {self.vocab_size_base} base tokens and {len(self.added_tokens_list)} added tokens>"
+
+
+class SentencePieceVocab:
+    def __init__(self, fname_tokenizer: Path, fname_added_tokens: Optional[Path]) -> None:
+        self.sentencepiece_tokenizer = SentencePieceProcessor(str(fname_tokenizer))
+        added_tokens: Dict[str, int]
+        if fname_added_tokens is not None:
+            added_tokens = json.load(open(fname_added_tokens, encoding="utf-8"))
         else:
+            added_tokens = {}
         vocab_size: int = self.sentencepiece_tokenizer.vocab_size()
-
         expected_ids = list(range(vocab_size, vocab_size + len(added_tokens)))
         actual_ids   = sorted(added_tokens.values())
         if expected_ids != actual_ids:
@@ -238,30 +271,9 @@ class SentencePieceVocab:
 
     def sentencepiece_tokens(self) -> Iterable[Tuple[bytes, float]]:
         tokenizer = self.sentencepiece_tokenizer
-        if self.vocabtype == "bpe":
-            from transformers.models.gpt2 import tokenization_gpt2
-            byte_encoder = tokenization_gpt2.bytes_to_unicode()
-            byte_decoder = {v: k for k, v in byte_encoder.items()}
-            for i, item in enumerate(tokenizer):
-                text: bytes
-                text = b''.join([x.to_bytes(1, byteorder='big') for x in [byte_decoder[y] for y in item]])
-                score: float = -i
-                yield text, score
-        else:
         for i in range(tokenizer.vocab_size()):
-                text: bytes
-                if tokenizer.is_unknown(i):
-                    text = " \u2047 ".encode("utf-8")
-                elif tokenizer.is_control(i):
-                    text = b""
-                elif tokenizer.is_byte(i):
             piece = tokenizer.id_to_piece(i)
-                    if len(piece) != 6:
+            text: bytes = piece.encode("utf-8")
-                        raise Exception(f"Invalid token: {piece}")
-                    byte_value = int(piece[3:-1], 16)
-                    text = struct.pack("B", byte_value)
-                else:
-                    text = tokenizer.id_to_piece(i).replace("\u2581", " ").encode("utf-8")
             score: float = tokenizer.get_score(i)
             yield text, score
 
@@ -278,7 +290,7 @@ class SentencePieceVocab:
         return f"<SentencePieceVocab with {self.vocab_size_base} base tokens and {len(self.added_tokens_list)} added tokens>"
 
 
-Vocab = Union[SentencePieceVocab]
+Vocab = Union[BpeVocab, SentencePieceVocab]
 
 
 def permute(weights: NDArray, n_head: int) -> NDArray:
@@ -679,7 +691,7 @@ def bounded_parallel_map(func: Callable[[In], Out], iterable: Iterable[In], conc
 
 def check_vocab_size(params: Params, vocab: Vocab) -> None:
     if params.n_vocab != vocab.vocab_size:
-        assert isinstance(vocab, SentencePieceVocab)
+        assert isinstance(vocab, BpeVocab) or isinstance(vocab, SentencePieceVocab)
         if params.n_vocab == vocab.vocab_size_base:
             print("Ignoring added_tokens.json since model matches vocab size without it.")
             vocab.added_tokens_list = []
@@ -853,7 +865,7 @@ def filter_and_sort_tensors(model: LazyModel) -> LazyModel:
     return {name: model[name] for name in TENSORS_LIST if name in model}
 
 
-def load_vocab(path: Path, vocabtype: Optional[str]) -> SentencePieceVocab:
+def load_vocab(path: Path, vocabtype: Optional[str]) -> Union[BpeVocab, SentencePieceVocab]:
     print(f"vocabtype: {vocabtype}")
     # Be extra-friendly and accept either a file or a directory.  Also, if it's
     # a directory, it might be the model directory, and tokenizer.model might
@@ -875,8 +887,12 @@ def load_vocab(path: Path, vocabtype: Optional[str]) -> SentencePieceVocab:
                 "if it's in another directory, pass the directory as --vocab-dir")
     added_tokens_path = path.parent / "added_tokens.json"
     print(f"Loading vocab file {path}")
-    return SentencePieceVocab(path, added_tokens_path if added_tokens_path.exists() else None,
+    if vocabtype == "bpe":
-                              vocabtype)
+        return BpeVocab(path, added_tokens_path if added_tokens_path.exists() else None)
+    elif vocabtype == "spm":
+        return SentencePieceVocab(path, added_tokens_path if added_tokens_path.exists() else None)
+    else:
+        raise ValueError(f"Unsupported vocabulary type {vocabtype}")
 
 
 def default_outfile(model_paths: List[Path], file_type: GGMLFileType) -> Path:

examples/main/main.cpp

@@ -307,7 +307,7 @@ int main(int argc, char ** argv) {
         auto console_ctrl_handler = +[](DWORD ctrl_type) -> BOOL {
             return (ctrl_type == CTRL_C_EVENT) ? (sigint_handler(SIGINT), true) : false;
         };
-        SetConsoleCtrlHandler(static_cast<PHANDLER_ROUTINE>(console_ctrl_handler), true);
+        SetConsoleCtrlHandler(reinterpret_cast<PHANDLER_ROUTINE>(console_ctrl_handler), true);
 #endif
 
         fprintf(stderr, "%s: interactive mode on.\n", __func__);