Align output with ChatGLM model

Add unit test for stream decoder
pull/817/head
lwh9346 2023-04-27 00:07:42 +08:00
parent d443215bea
commit 430224bf13
2 changed files with 96 additions and 4 deletions

View File

@ -2,12 +2,12 @@ import os
from transformers import AutoTokenizer, AutoModel from transformers import AutoTokenizer, AutoModel
import signal import signal
import platform import platform
from stream_utils import SPStreamDecoder from stream_utils import ChatGLMStreamDecoder
tokenizer = AutoTokenizer.from_pretrained( tokenizer = AutoTokenizer.from_pretrained(
"THUDM/chatglm-6b", trust_remote_code=True) "THUDM/chatglm-6b", trust_remote_code=True)
stream_decoder = SPStreamDecoder(tokenizer.sp_tokenizer.text_tokenizer.sp) stream_decoder = ChatGLMStreamDecoder(tokenizer.sp_tokenizer.text_tokenizer.sp)
model = AutoModel.from_pretrained( model = AutoModel.from_pretrained(
"THUDM/chatglm-6b", trust_remote_code=True).half().cuda() "THUDM/chatglm-6b", trust_remote_code=True).half().cuda()
model = model.eval() model = model.eval()

View File

@ -1,5 +1,7 @@
import sentencepiece as spm import sentencepiece as spm
from typing import Tuple from typing import Tuple
import re
import unittest
# python implantation of https://github.com/google/sentencepiece/blob/master/src/sentencepiece_processor.cc # python implantation of https://github.com/google/sentencepiece/blob/master/src/sentencepiece_processor.cc
@ -24,6 +26,7 @@ def DecodeSentencePiece(piece: str, id: int, is_bos_ws: bool, sp: spm.SentencePi
if is_bos_ws and (add_dummy_prefix or remove_extra_whitespaces): if is_bos_ws and (add_dummy_prefix or remove_extra_whitespaces):
t = piece.removeprefix(SPStreamDecoder.SpaceSymbol) t = piece.removeprefix(SPStreamDecoder.SpaceSymbol)
has_bos_ws = t != piece has_bos_ws = t != piece
piece = t
# if we are removing extra whitespace, we remove all leading whitespace # if we are removing extra whitespace, we remove all leading whitespace
if remove_extra_whitespaces: if remove_extra_whitespaces:
has_bos_ws = False has_bos_ws = False
@ -67,17 +70,21 @@ class SPStreamDecoder:
self._nothing_decoded = True self._nothing_decoded = True
self._ids = [] self._ids = []
self._decoded = "" self._decoded = ""
self._ending = False
self.remove_extra_whitespaces = remove_extra_whitespaces self.remove_extra_whitespaces = remove_extra_whitespaces
self.add_dummy_prefix = add_dummy_prefix self.add_dummy_prefix = add_dummy_prefix
def put(self, ids: list[int]) -> None: def put(self, ids: list[int]) -> None:
self._ending = False
self._ids += ids self._ids += ids
self._decode(eos=False) self._decode(eos=False)
def end(self) -> None: def end(self) -> None:
self._decode(eos=True) self._decode(eos=True)
self._is_bos_ws = True
self._bos_ws_seen = False self._bos_ws_seen = False
self._nothing_decoded = True self._nothing_decoded = True
self._ending = True
self._ids = [] self._ids = []
def _decode(self, eos=False) -> None: def _decode(self, eos=False) -> None:
@ -88,7 +95,7 @@ class SPStreamDecoder:
if not self._sp.IsByte(self._ids[i]): if not self._sp.IsByte(self._ids[i]):
self._decoded += ProcessBytePieces(byte_pieces) self._decoded += ProcessBytePieces(byte_pieces)
consumed += len(byte_pieces) consumed += len(byte_pieces)
if consumed > 0: if len(self._decoded) > 0:
self._nothing_decoded = False self._nothing_decoded = False
byte_pieces = [] byte_pieces = []
# if we have seen a bos_ws token or any non-empty token # if we have seen a bos_ws token or any non-empty token
@ -98,7 +105,7 @@ class SPStreamDecoder:
piece, self._ids[i], self._is_bos_ws, self._sp) piece, self._ids[i], self._is_bos_ws, self._sp)
self._decoded += decoded self._decoded += decoded
consumed += 1 consumed += 1
if consumed > 0: if len(self._decoded) > 0:
self._nothing_decoded = False self._nothing_decoded = False
else: else:
byte_pieces.append(piece) byte_pieces.append(piece)
@ -111,3 +118,88 @@ class SPStreamDecoder:
t = self._decoded t = self._decoded
self._decoded = "" self._decoded = ""
return t return t
class ChatGLMStreamDecoder(SPStreamDecoder):
def get(self) -> str:
# if prefix of special tokens found, wait till it's impossible or end of decode
if "[" in self._decoded and len(self._decoded)-self._decoded.index("[") < 8 and not self._ending:
return ""
if "<" in self._decoded and len(self._decoded)-self._decoded.index("<") < 12 and not self._ending:
return ""
self._ending = False
t = self._decoded
self._decoded = ""
t = t.replace("<n>", "\n")
t = t.replace("[[训练时间]]", "2023年")
punkts = [
[",", ""],
["!", ""],
[":", ""],
[";", ""],
["\?", ""],
]
for item in punkts:
t = re.sub(r"([\u4e00-\u9fff])%s" % item[0], r"\1%s" % item[1], t)
t = re.sub(r"%s([\u4e00-\u9fff])" % item[0], r"%s\1" % item[1], t)
# for i in range(max_len, 1, -1):
# t = t.replace(f"<|blank_{i}|>", " " * i)
for blank_token in re.findall(r"<\|blank_\d+\|>", t):
t = t.replace(blank_token, " " *
int(re.search(r"\d+", blank_token)[0]))
return t
class ChatGLMStreamDecoderTest(unittest.TestCase):
def test_ChatGLM_StreamDecoder(self):
from transformers import AutoTokenizer, AutoModel
test_strings = [
"你好👋", # multi-byte encoding
"Hello this is ChatGLM!", # normal text
"你好👋 This is ChatGLM!", # multi-byte encoding with tail
"!?.,!?。,", # punctuations
"A\nB", # "<n>" -> "\n"
"[[训练时间]]", # training time token
"[[训练时间]123", # broken training time token
"1 1", # blank token. Note: It's hard to match the results of strip(), so add leading and tailing "1"
"<|blank_8|123", # broken blank token
]
tokenizer = AutoTokenizer.from_pretrained(
"THUDM/chatglm-6b", trust_remote_code=True)
model = AutoModel.from_pretrained(
"THUDM/chatglm-6b", trust_remote_code=True).half().cuda()
model = model.eval()
encoded_ids = [tokenizer(x)['input_ids'] for x in test_strings]
stream_decoder = ChatGLMStreamDecoder(
tokenizer.sp_tokenizer.text_tokenizer.sp)
# original output
expected_outputs = [model.process_response(
tokenizer.decode(x)) for x in encoded_ids]
# decode token by token
decoded_strings_stream_token_by_token = [None for _ in test_strings]
for i in range(len(test_strings)):
res = []
for t in encoded_ids[i]:
stream_decoder.put([t])
res.append(stream_decoder.get())
stream_decoder.end()
res.append(stream_decoder.get())
res = "".join(res)
decoded_strings_stream_token_by_token[i] = res
# decode all at once
decoded_strings_stream = [None for _ in test_strings]
for i in range(len(test_strings)):
stream_decoder.put(encoded_ids[i])
stream_decoder.end()
decoded_strings_stream[i] = stream_decoder.get()
for i in range(len(test_strings)):
print(
f"Stream decoder test{i}: expected: '{expected_outputs[i]}', token_by_token: '{decoded_strings_stream_token_by_token[i]}', all at once: '{decoded_strings_stream[i]}'")
self.assertEqual(
expected_outputs[i], decoded_strings_stream_token_by_token[i])
self.assertEqual(expected_outputs[i], decoded_strings_stream[i])
if __name__ == "__main__":
unittest.main()