Work on the BPE tokenizer (#3252)

* Work on the BPE tokenizer Tokenizer tests work for Falcon-7B * Try to fix build problem * Fix debug assertion failure * Fix MSVC Unicode BOM problem * Cleanup and an improvement * Fix compiler warning * Cleanup * Test doesn't work over the full range of Unicodes * Update .gitignore and Makefile * Another Makefile rule * Testing Aquila * Moving byte decoding back to `token_to_piece` ... ... because everyone is using it. * Guarding some unusable code pathes * Streamlining code and adding some more assertions Important change: I'm classifying added tokens as control tokens now for BPE. * Adding a comment * Adding another assertion * Fixed vocabulary guarding assertions * Fix PR for recent change * Fix PR for recent change * Fix for compiler warning * Fix PR for recent change * Fix PR for recent change * Fix PR for recent change * Fix for compiler warning * Fixes for more compiler warnings * Remove unused code * Fix initialization of static maps * Add scores and token types back, adapt gptneox * Update llama.cpp Co-authored-by: Georgi Gerganov <[email protected]> * Update unicode.h Co-authored-by: Georgi Gerganov <[email protected]> * Update unicode.h Co-authored-by: Georgi Gerganov <[email protected]> * Ported Starcoder and added some assertions * Fix coding style * Apply @jploski 's fix for missing tokens --------- Co-authored-by: Georgi Gerganov <[email protected]>
ggerganov · Oct 3, 2023 · ff5a3f0 · ff5a3f0
1 parent 1c84003
commit ff5a3f0
Show file tree

Hide file tree

Showing 15 changed files with 852 additions and 227 deletions.
diff --git a/.gitignore b/.gitignore
@@ -91,4 +91,5 @@ tests/test-quantize-perf
 tests/test-sampling
 tests/test-tokenizer-0-llama
 tests/test-tokenizer-0-falcon
-tests/test-tokenizer-1
+tests/test-tokenizer-1-llama
+tests/test-tokenizer-1-bpe
diff --git a/Makefile b/Makefile
@@ -2,7 +2,7 @@
 BUILD_TARGETS = main quantize quantize-stats perplexity embedding vdot q8dot train-text-from-scratch convert-llama2c-to-ggml simple batched save-load-state server embd-input-test gguf llama-bench baby-llama beam-search speculative infill benchmark-matmult parallel finetune export-lora tests/test-c.o
 
 # Binaries only useful for tests
-TEST_TARGETS = tests/test-llama-grammar tests/test-grammar-parser tests/test-double-float tests/test-grad0 tests/test-opt tests/test-quantize-fns tests/test-quantize-perf tests/test-sampling tests/test-tokenizer-0-llama tests/test-tokenizer-0-falcon tests/test-tokenizer-1-llama
+TEST_TARGETS = tests/test-llama-grammar tests/test-grammar-parser tests/test-double-float tests/test-grad0 tests/test-opt tests/test-quantize-fns tests/test-quantize-perf tests/test-sampling tests/test-tokenizer-0-llama tests/test-tokenizer-0-falcon tests/test-tokenizer-1-llama tests/test-tokenizer-1-bpe
 
 # Code coverage output files
 COV_TARGETS = *.gcno tests/*.gcno *.gcda tests/*.gcda *.gcov tests/*.gcov lcov-report gcovr-report
@@ -62,9 +62,11 @@ test: $(TEST_TARGETS)
 		if [ "$$test_target" = "tests/test-tokenizer-0-llama" ]; then \
 			./$$test_target $(CURDIR)/models/ggml-vocab-llama.gguf; \
 		elif [ "$$test_target" = "tests/test-tokenizer-0-falcon" ]; then \
-			continue; \
+			./$$test_target $(CURDIR)/models/ggml-vocab-falcon.gguf; \
 		elif [ "$$test_target" = "tests/test-tokenizer-1-llama" ]; then \
 			continue; \
+		elif [ "$$test_target" = "tests/test-tokenizer-1-bpe" ]; then \
+			continue; \
 		else \
 			echo "Running test $$test_target..."; \
 			./$$test_target; \
@@ -670,6 +672,9 @@ tests/test-tokenizer-0-falcon: tests/test-tokenizer-0-falcon.cpp build-info.h gg
 tests/test-tokenizer-0-llama: tests/test-tokenizer-0-llama.cpp build-info.h ggml.o llama.o common.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
+tests/test-tokenizer-1-bpe: tests/test-tokenizer-1-bpe.cpp build-info.h ggml.o llama.o common.o $(OBJS)
+	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
+
 tests/test-tokenizer-1-llama: tests/test-tokenizer-1-llama.cpp build-info.h ggml.o llama.o common.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 

diff --git a/common/common.cpp b/common/common.cpp
@@ -923,6 +923,7 @@ std::string llama_detokenize_bpe(llama_context * ctx, const std::vector<llama_to
         result += piece;
     }
 
+    // NOTE: the original tokenizer decodes bytes after collecting the pieces.
     return result;
 }
 

diff --git a/convert-falcon-hf-to-gguf.py b/convert-falcon-hf-to-gguf.py
@@ -20,28 +20,6 @@
 import gguf
 
 
-def bytes_to_unicode():
-    # ref: https://github.com/openai/gpt-2/blob/master/src/encoder.py
-    """
-    Returns list of utf-8 byte and a corresponding list of unicode strings.
-    The reversible bpe codes work on unicode strings.
-    This means you need a large # of unicode characters in your vocab if you want to avoid UNKs.
-    When you're at something like a 10B token dataset you end up needing around 5K for decent coverage.
-    This is a significant percentage of your normal, say, 32K bpe vocab.
-    To avoid that, we want lookup tables between utf-8 bytes and unicode strings.
-    And avoids mapping to whitespace/control characters the bpe code barfs on.
-    """
-    bs = list(range(ord("!"), ord("~")+1))+list(range(ord("¡"), ord("¬")+1))+list(range(ord("®"), ord("ÿ")+1))
-    cs = bs[:]
-    n = 0
-    for b in range(2**8):
-        if b not in bs:
-            bs.append(b)
-            cs.append(2**8+n)
-            n += 1
-    return dict(zip(bs, (chr(n) for n in cs)))
-
-
 def count_model_parts(dir_model: Path) -> int:
     num_parts = 0
     for filename in os.listdir(dir_model):
@@ -133,6 +111,8 @@ def parse_args() -> argparse.Namespace:
 print("gguf: get tokenizer metadata")
 
 tokens: list[bytearray] = []
+scores: list[float] = []
+toktypes: list[int] = []
 
 # gpt2 tokenizer
 gguf_writer.add_tokenizer_model("gpt2")
@@ -148,28 +128,15 @@ def parse_args() -> argparse.Namespace:
 assert max(tokenizer.vocab.values()) < vocab_size
 
 reverse_vocab = {id: encoded_tok for encoded_tok, id in tokenizer.vocab.items()}
-byte_encoder = bytes_to_unicode()
-byte_decoder = {v: k for k, v in byte_encoder.items()}
 
 for i in range(vocab_size):
-    if i in reverse_vocab:
-        try:
-            text = bytearray([byte_decoder[c] for c in reverse_vocab[i]])
-        except KeyError:
-            text = bytearray()
-            for c in reverse_vocab[i]:
-                if ord(c) < 256:  # single byte character
-                    text.append(byte_decoder[ord(c)])
-                else:  # multibyte special token character
-                    text.extend(c.encode('utf-8'))
-    else:
-        print(f"Key {i} not in tokenizer vocabulary. Padding with an arbitrary token.")
-        pad_token = f"[PAD{i}]".encode("utf8")
-        text = bytearray(pad_token)
-
-    tokens.append(text)
+    tokens.append(reverse_vocab[i])
+    scores.append(0.0) # dummy
+    toktypes.append(gguf.TokenType.NORMAL)
 
 gguf_writer.add_token_list(tokens)
+gguf_writer.add_token_scores(scores)
+gguf_writer.add_token_types(toktypes)
 
 special_vocab = gguf.SpecialVocab(dir_model, load_merges = True)
 special_vocab.add_to_gguf(gguf_writer)

diff --git a/convert-gptneox-hf-to-gguf.py b/convert-gptneox-hf-to-gguf.py
@@ -19,29 +19,6 @@
     sys.path.insert(1, str(Path(__file__).parent / 'gguf-py' / 'gguf'))
 import gguf
 
-# ref: https://github.com/openai/gpt-2/blob/master/src/encoder.py
-
-
-def bytes_to_unicode():
-    """
-    Returns list of utf-8 byte and a corresponding list of unicode strings.
-    The reversible bpe codes work on unicode strings.
-    This means you need a large # of unicode characters in your vocab if you want to avoid UNKs.
-    When you're at something like a 10B token dataset you end up needing around 5K for decent coverage.
-    This is a significant percentage of your normal, say, 32K bpe vocab.
-    To avoid that, we want lookup tables between utf-8 bytes and unicode strings.
-    And avoids mapping to whitespace/control characters the bpe code barfs on.
-    """
-    bs = list(range(ord("!"), ord("~")+1))+list(range(ord("¡"), ord("¬")+1))+list(range(ord("®"), ord("ÿ")+1))
-    cs = bs[:]
-    n = 0
-    for b in range(2**8):
-        if b not in bs:
-            bs.append(b)
-            cs.append(2**8+n)
-            n += 1
-    return dict(zip(bs, (chr(n) for n in cs)))
-
 
 def count_model_parts(dir_model: Path) -> int:
     num_parts = 0
@@ -130,6 +107,8 @@ def parse_args() -> argparse.Namespace:
 print("gguf: get tokenizer metadata")
 
 tokens: list[bytearray] = []
+scores: list[float] = []
+toktypes: list[int] = []
 
 # gpt2 tokenizer
 gguf_writer.add_tokenizer_model("gpt2")
@@ -145,28 +124,15 @@ def parse_args() -> argparse.Namespace:
 assert max(tokenizer.vocab.values()) < vocab_size
 
 reverse_vocab = {id: encoded_tok for encoded_tok, id in tokenizer.vocab.items()}
-byte_encoder = bytes_to_unicode()
-byte_decoder = {v: k for k, v in byte_encoder.items()}
 
 for i in range(vocab_size):
-    if i in reverse_vocab:
-        try:
-            text = bytearray([byte_decoder[c] for c in reverse_vocab[i]])
-        except KeyError:
-            text = bytearray()
-            for c in reverse_vocab[i]:
-                if ord(c) < 256:  # single byte character
-                    text.append(byte_decoder[ord(c)])
-                else:  # multibyte special token character
-                    text.extend(c.encode('utf-8'))
-    else:
-        print(f"Key {i} not in tokenizer vocabulary. Padding with an arbitrary token.")
-        pad_token = f"[PAD{i}]".encode("utf8")
-        text = bytearray(pad_token)
-
-    tokens.append(text)
+    tokens.append(reverse_vocab[i] if i in reverse_vocab else f"[PAD{i}]")
+    scores.append(0.0) # dummy
+    toktypes.append(gguf.TokenType.NORMAL)
 
 gguf_writer.add_token_list(tokens)
+gguf_writer.add_token_scores(scores)
+gguf_writer.add_token_types(toktypes)
 
 special_vocab = gguf.SpecialVocab(dir_model, load_merges = True)
 special_vocab.add_to_gguf(gguf_writer)

diff --git a/convert-starcoder-hf-to-gguf.py b/convert-starcoder-hf-to-gguf.py
@@ -20,28 +20,6 @@
 import gguf
 
 
-def bytes_to_unicode():
-    # ref: https://github.com/openai/gpt-2/blob/master/src/encoder.py
-    """
-    Returns list of utf-8 byte and a corresponding list of unicode strings.
-    The reversible bpe codes work on unicode strings.
-    This means you need a large # of unicode characters in your vocab if you want to avoid UNKs.
-    When you're at something like a 10B token dataset you end up needing around 5K for decent coverage.
-    This is a significant percentage of your normal, say, 32K bpe vocab.
-    To avoid that, we want lookup tables between utf-8 bytes and unicode strings.
-    And avoids mapping to whitespace/control characters the bpe code barfs on.
-    """
-    bs = list(range(ord("!"), ord("~")+1))+list(range(ord("¡"), ord("¬")+1))+list(range(ord("®"), ord("ÿ")+1))
-    cs = bs[:]
-    n = 0
-    for b in range(2**8):
-        if b not in bs:
-            bs.append(b)
-            cs.append(2**8+n)
-            n += 1
-    return dict(zip(bs, (chr(n) for n in cs)))
-
-
 def count_model_parts(dir_model: Path) -> int:
     num_parts = 0
     for filename in os.listdir(dir_model):
@@ -117,6 +95,8 @@ def parse_args() -> argparse.Namespace:
 print("gguf: get tokenizer metadata")
 
 tokens: list[bytearray] = []
+scores: list[float] = []
+toktypes: list[int] = []
 
 # gpt2 tokenizer
 gguf_writer.add_tokenizer_model("gpt2")
@@ -132,28 +112,15 @@ def parse_args() -> argparse.Namespace:
 assert max(tokenizer.vocab.values()) < vocab_size
 
 reverse_vocab = {id: encoded_tok for encoded_tok, id in tokenizer.vocab.items()}
-byte_encoder = bytes_to_unicode()
-byte_decoder = {v: k for k, v in byte_encoder.items()}
 
 for i in range(vocab_size):
-    if i in reverse_vocab:
-        try:
-            text = bytearray([byte_decoder[c] for c in reverse_vocab[i]])
-        except KeyError:
-            text = bytearray()
-            for c in reverse_vocab[i]:
-                if ord(c) < 256:  # single byte character
-                    text.append(byte_decoder[ord(c)])
-                else:  # multibyte special token character
-                    text.extend(c.encode('utf-8'))
-    else:
-        print(f"Key {i} not in tokenizer vocabulary. Padding with an arbitrary token.")
-        pad_token = f"[PAD{i}]".encode("utf8")
-        text = bytearray(pad_token)
-
-    tokens.append(text)
+    tokens.append(reverse_vocab[i] if i in reverse_vocab else f"[PAD{i}]")
+    scores.append(0.0) # dummy
+    toktypes.append(gguf.TokenType.NORMAL)
 
 gguf_writer.add_token_list(tokens)
+gguf_writer.add_token_scores(scores)
+gguf_writer.add_token_types(toktypes)
 
 special_vocab = gguf.SpecialVocab(dir_model, load_merges = True)
 special_vocab.add_to_gguf(gguf_writer)

diff --git a/convert.py b/convert.py
@@ -338,29 +338,15 @@ def __init__(self, fname_tokenizer: Path, fname_added_tokens: Path | None) -> No
     def bpe_tokens(self) -> Iterable[tuple[bytes, float, gguf.TokenType]]:
         tokenizer = self.bpe_tokenizer
         from transformers.models.gpt2 import tokenization_gpt2  # type: ignore[import]
-        byte_encoder = tokenization_gpt2.bytes_to_unicode()
-        byte_decoder = {v: k for k, v in byte_encoder.items()}
-        score = 0.0
-        for i, item in enumerate(tokenizer):
-            text: bytes = item.encode("utf-8")
-            # FIXME: These shouldn't be hardcoded, but it's probably better than the current behavior?
-            if i <= 258 and text.startswith(b'<') and text.endswith(b'>'):
-                if i == 0 and text == b'<unk>':
-                    toktype = gguf.TokenType.UNKNOWN
-                elif i == 1 or i == 2:
-                    toktype = gguf.TokenType.CONTROL
-                elif i >= 3 and text.startswith(b'<0x'):
-                    toktype = gguf.TokenType.BYTE
-                else:
-                    toktype = gguf.TokenType.NORMAL
-            else:
-                toktype = gguf.TokenType.NORMAL
-            yield text, score, toktype
+        reverse_vocab = {id: encoded_tok for encoded_tok, id in tokenizer.items()}
+
+        for i, _ in enumerate(tokenizer):
+            yield reverse_vocab[i], 0.0, gguf.TokenType.NORMAL
 
     def added_tokens(self) -> Iterable[tuple[bytes, float, gguf.TokenType]]:
         for text in self.added_tokens_list:
             score = -1000.0
-            yield text.encode("utf-8"), score, gguf.TokenType.USER_DEFINED
+            yield text.encode("utf-8"), score, gguf.TokenType.CONTROL
 
     def all_tokens(self) -> Iterable[tuple[bytes, float, gguf.TokenType]]:
         yield from self.bpe_tokens()