如题

可以看到字符都是看不见的，当选中时，字符才显示出来

加载了本地马保国文本数据后报错了

可否实现将两个模型（chatglm和text2vec）加载到不同的GPU上呢，我发现自己的GPU（12G显存，实际可用11G多点）运行时显存会溢出，想问下将两个模型加载到不同GPU上是否可以实现

请问联网功能是如何实现的呢？

2023-04-20 13:34:51,654 [INFO] [loader.py:54] Loading faiss with AVX2 support.
2023-04-20 13:34:52,051 [INFO] [loader.py:56] Successfully loaded faiss with AVX2 support.
Error in faiss::FileIOReader::FileIOReader(const char*) at /project/faiss/faiss/impl/io.cpp:67: Error: 'f' failed: could not open cache/financial_research_reports/index.faiss for reading: No such file or directory

Resource [93mpunkt[om not found.Please use the NLTK Downloader to obtain theresource:
[31m>>> import nltk>>> nltk.download(' punkt')[omFor more information see: https://www.nltk.org/data.html
NLTK's dat
all"), or just tra and no gra
Attempted to load [93mtokenizers/punkt/PY3/english.pickle[om
Searched in:/home/horace/nltk data!/home/horace/.conda/envs/chatglm/nltk data!- '/home/horace/.conda/envs/chatglm/share/nltk data'/home/horace/.conda/envs/chatglm/lib/nltk data'
'/usr/share/nltk data'
'/usr/local/share/nltk data
‘/usr/lib/nltk data’
‘/usr/local/lib/nltk data’

终端中提示我import nltk后使用nltk.download('punkt')，但执行这条命令会失败。查询资料后去官网下载，解压后放到上述提到的指定位置，再运行main.py，发现还是出现这个错误。不知道怎么解决

无法使用“加载知识库”功能

启动服务后只能使用模型推理；

尝试点击：“加载知识库”，报错：”初始化知识库未成功加载“

报错代码：

Error in faiss::FileIOReader::FileIOReader(const char*) at /project/faiss/faiss/impl/io.cpp:67: Error: 'f' failed: could not open cache/index.faiss for reading: No such file or directory

查看源码，并没有发现cache目录；

文件上传功能

左下角的“上传文件到知识库”，文件上传后会保存到docs目录下，但如果是中文名的文件，上传到linux上会乱码；

文件上传后不知道如何使用，仍然是无法加载知识库；

单卡多GPU 报gradio port被占用

Explicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision.
Explicitly passing a `revision` is encouraged when loading a configuration with custom code to ensure no malicious code has been contributed in a newer revision.
Explicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision.
No compiled kernel found.
Compiling kernels : /root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int8/2cc27c946766f6ebd1504001b5d776a27c4d0d0b/quantization_kernels_parallel.c
Compiling gcc -O3 -fPIC -pthread -fopenmp -std=c99 /root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int8/2cc27c946766f6ebd1504001b5d776a27c4d0d0b/quantization_kernels_parallel.c -shared -o /root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int8/2cc27c946766f6ebd1504001b5d776a27c4d0d0b/quantization_kernels_parallel.so
Load kernel : /root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int8/2cc27c946766f6ebd1504001b5d776a27c4d0d0b/quantization_kernels_parallel.so
Setting CPU quantization kernel threads to 6
Using quantization cache
Applying quantization to glm layers
2023-04-20 17:41:52,801 [WARNING] [SentenceTransformer.py:805] No sentence-transformers model found with name /root/.cache/torch/sentence_transformers/GanymedeNil_text2vec-large-chinese. Creating a new one with MEAN pooling.
Traceback (most recent call last):
  File "/home/heyiheng/work/Chinese-LangChain/main.py", line 27, in <module>
    application.source_service.init_source_vector()
  File "/home/heyiheng/work/Chinese-LangChain/clc/source_service.py", line 37, in init_source_vector
    print(doc)
UnicodeEncodeError: `'utf-8' codec can't encode characters in position 1-2: surrogates not allowed`

一个我无法解决的报错，有空的话请帮忙看一下

RT

两个 中文维基百科 的网盘链接里都缺少 index.pkl呀？
如果要自己重新生成的话。。那提供index.faiss有什么意义？

Traceback (most recent call last):
File "/root/Chinese-LangChain/main.py", line 26, in
application = LangChainApplication(config)
File "/root/Chinese-LangChain/clc/langchain_application.py", line 24, in init
self.llm_service.load_model(model_name_or_path=self.config.llm_model_name)
File "/root/Chinese-LangChain/clc/gpt_service.py", line 55, in load_model
self.tokenizer = AutoTokenizer.from_pretrained(
File "/root/anaconda3/envs/langchain/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py", line 721, in from_pretrained
raise ValueError(
ValueError: Unrecognized configuration class <class 'transformers_modules.THUDM.chatglm-6b-int4-qe.977d9df4cfae6b7a756e07698483872c5c070eee.configuration_chatglm.ChatGLMConfig'> to build an AutoTokenizer.
Model type should be one of AlbertConfig, AlignConfig, BartConfig, BertConfig, BertGenerationConfig, BigBirdConfig, BigBirdPegasusConfig, BioGptConfig, BlenderbotConfig, BlenderbotSmallConfig, BlipConfig, Blip2Config, BloomConfig, BridgeTowerConfig, CamembertConfig, CanineConfig, ChineseCLIPConfig, ClapConfig, CLIPConfig, CLIPSegConfig, CodeGenConfig, ConvBertConfig, CpmAntConfig, CTRLConfig, Data2VecTextConfig, DebertaConfig, DebertaV2Config, DistilBertConfig, DPRConfig, ElectraConfig, ErnieConfig, ErnieMConfig, EsmConfig, FlaubertConfig, FNetConfig, FSMTConfig, FunnelConfig, GitConfig, GPT2Config, GPT2Config, GPTBigCodeConfig, GPTNeoConfig, GPTNeoXConfig, GPTNeoXJapaneseConfig, GPTJConfig, GPTSanJapaneseConfig, GroupViTConfig, HubertConfig, IBertConfig, JukeboxConfig, LayoutLMConfig, LayoutLMv2Config, LayoutLMv3Config, LEDConfig, LiltConfig, LlamaConfig, LongformerConfig, LongT5Config, LukeConfig, LxmertConfig, M2M100Config, MarianConfig, MBartConfig, MegaConfig, MegatronBertConfig, MgpstrConfig, MobileBertConfig, MPNetConfig, MT5Config, MvpConfig, NezhaConfig, NllbMoeConfig, NystromformerConfig, OneFormerConfig, OpenAIGPTConfig, OPTConfig, OwlViTConfig, PegasusConfig, PegasusXConfig, PerceiverConfig, Pix2StructConfig, PLBartConfig, ProphetNetConfig, QDQBertConfig, RagConfig, RealmConfig, ReformerConfig, RemBertConfig, RetriBertConfig, RobertaConfig, RobertaPreLayerNormConfig, RoCBertConfig, RoFormerConfig, RwkvConfig, Speech2TextConfig, Speech2Text2Config, SpeechT5Config, SplinterConfig, SqueezeBertConfig, SwitchTransformersConfig, T5Config, TapasConfig, TransfoXLConfig, ViltConfig, VisualBertConfig, Wav2Vec2Config, Wav2Vec2ConformerConfig, WhisperConfig, XCLIPConfig, XGLMConfig, XLMConfig, XLMProphetNetConfig, XLMRobertaConfig, XLMRobertaXLConfig, XLNetConfig, XmodConfig, YosoConfig.

cannot import name 'RetrievalQA' from 'langchain.chains'

2023-05-31 11:38:31.955 Uncaught app exception
Traceback (most recent call last):
File "/usr/local/python3/lib/python3.9/shutil.py", line 815, in move
os.rename(src, real_dst)
FileNotFoundError: [Errno 2] No such file or directory

希望可以加上Stream模式的输出，用户体验更好

当在页面上提问时发生CUDA运行期错误。我的环境：CentOS 7.9+P100单卡+NVIDIA-SMI 460.91.03 Driver Version: 460.91.03 CUDA Version: 11.2+Python 3.10.9，Torch试过了2.0.1和1.8.0都是同样的错。

2023-05-21 23:08:31,755 [WARNING] [logging.py:295] The dtype of attention mask (torch.int64) is not bool
Traceback (most recent call last):
File "/opt/Python/lib/python3.10/site-packages/gradio/routes.py", line 422, in run_predict
output = await app.get_blocks().process_api(
File "/opt/Python/lib/python3.10/site-packages/gradio/blocks.py", line 1323, in process_api
result = await self.call_function(
File "/opt/Python/lib/python3.10/site-packages/gradio/blocks.py", line 1051, in call_function
prediction = await anyio.to_thread.run_sync(
File "/opt/Python/lib/python3.10/site-packages/anyio/to_thread.py", line 31, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "/opt/Python/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 937, in run_sync_in_worker_thread
return await future
File "/opt/Python/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 867, in run
result = context.run(func, *args)
File "/root/Chinese-LangChain/main.py", line 82, in predict
result = application.get_llm_answer(query=input, web_content=web_content)
File "/root/Chinese-LangChain/clc/langchain_application.py", line 85, in get_llm_answer
result = self.llm_service._call(prompt)
File "/root/Chinese-LangChain/clc/gpt_service.py", line 41, in _call
response, _ = self.model.chat(
File "/opt/Python/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "/root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4-qe/977d9df4cfae6b7a756e07698483872c5c070eee/modeling_chatglm.py", line 1253, in chat
outputs = self.generate(**inputs, **gen_kwargs)
File "/opt/Python/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "/opt/Python/lib/python3.10/site-packages/transformers/generation/utils.py", line 1565, in generate
return self.sample(
File "/opt/Python/lib/python3.10/site-packages/transformers/generation/utils.py", line 2612, in sample
outputs = self(
File "/opt/Python/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4-qe/977d9df4cfae6b7a756e07698483872c5c070eee/modeling_chatglm.py", line 1158, in forward
transformer_outputs = self.transformer(
File "/opt/Python/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4-qe/977d9df4cfae6b7a756e07698483872c5c070eee/modeling_chatglm.py", line 905, in forward
inputs_embeds = self.word_embeddings(input_ids)
File "/opt/Python/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4-qe/977d9df4cfae6b7a756e07698483872c5c070eee/quantization.py", line 384, in forward
original_weight = extract_weight_to_half(weight=self.weight, scale_list=self.weight_scale, source_bit_width=self.weight_bit_width)
File "/root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4-qe/977d9df4cfae6b7a756e07698483872c5c070eee/quantization.py", line 239, in extract_weight_to_half
func(
File "/opt/Python/lib/python3.10/site-packages/cpm_kernels/kernels/base.py", line 48, in call
func = self._prepare_func()
File "/opt/Python/lib/python3.10/site-packages/cpm_kernels/kernels/base.py", line 40, in _prepare_func
self._module.get_module(), self._func_name
File "/opt/Python/lib/python3.10/site-packages/cpm_kernels/kernels/base.py", line 24, in get_module
self._module[curr_device] = cuda.cuModuleLoadData(self._code)
File "/opt/Python/lib/python3.10/site-packages/cpm_kernels/library/base.py", line 94, in wrapper
return f(*args, **kwargs)
File "/opt/Python/lib/python3.10/site-packages/cpm_kernels/library/cuda.py", line 233, in cuModuleLoadData
checkCUStatus(cuda.cuModuleLoadData(ctypes.byref(module), data))
File "/opt/Python/lib/python3.10/site-packages/cpm_kernels/library/cuda.py", line 216, in checkCUStatus
raise RuntimeError("CUDA Error: %s" % cuGetErrorString(error))
RuntimeError: CUDA Error: no kernel image is available for execution on the device

操作系统版本：Ubuntu 20.04 LTS
python版本：3.10.9
transformers安装的时候没有指定版本，安装的是最新4.24.0版
(base) root@st-desktop:~/xxy/Chinese-LangChain# python -m pip list |grep transformers
sentence-transformers 2.2.2
transformers 4.24.0

部分报错信息如下：
│ │
│ /root/anaconda3/lib/python3.10/importlib/init.py:126 in import_module │
│ │
│ 123 │ │ │ if character != '.': │
│ 124 │ │ │ │ break │
│ 125 │ │ │ level += 1 │
│ ❱ 126 │ return _bootstrap._gcd_import(name[level:], package, level) │
│ 127 │
│ 128 │
│ 129 _RELOADING = {} │
│ :1050 in _gcd_import │
│ :1027 in _find_and_load │
│ :1006 in _find_and_load_unlocked │
│ :688 in _load_unlocked │
│ :883 in exec_module │
│ :241 in _call_with_frames_removed │
│ │
│ /root/.cache/huggingface/modules/transformers_modules/THUDM/chatglm-6b-int4-qe/977d9df4cfae6b7a7 │
│ 56e07698483872c5c070eee/modeling_chatglm.py:30 in │
│ │
│ 27 ) │
│ 28 from transformers.modeling_utils import PreTrainedModel │
│ 29 from transformers.utils import logging │
│ ❱ 30 from transformers.generation.logits_process import LogitsProcessor │
│ 31 from transformers.generation.utils import LogitsProcessorList, StoppingCriteriaList, Gen │
│ 32 │
│ 33 from .configuration_chatglm import ChatGLMConfig │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
ModuleNotFoundError: No module named 'transformers.generation'

请教大家

另外，是否应该列出当前已经传过的所有知识库，方便随时加载切换

如题，期待您的回复！

python main.py时报错：
│ D:\miniconda3\envs\langcn\lib\site-packages\unstructured\partition\pdf.py:5 in │
│ │
│ 2 from io import StringIO │
│ 3 from typing import BinaryIO, List, Optional, cast │
│ 4 │
│ ❱ 5 from pdfminer.pdfpage import PDFPage, PDFTextExtractionNotAllowed │
│ 6 from pdfminer.utils import open_filename │
│ 7 │
│ 8 from unstructured.documents.elements import Element, ElementMetadata, PageBreak │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
ModuleNotFoundError: No module named 'pdfminer'

库文件调整为
unstructured[local-inference]
不报错了

设置embedding_model_name = './text2vec-large-chinese'，运行create_knowledge.py代码，提示
No sentence-transformers model found with name ./text2vec-large-chinese Creating a new one with MEAN pooling.
text2vec-large-chinese模型文件是从huggingface(https://huggingface.co/GanymedeNil/text2vec-large-chinese/tree/main)上下载的；
请问是什么原因呢？

C:\Users\admin\Desktop\Chinese-LangChain-master>py main.py
Traceback (most recent call last):
File "C:\Users\admin\Desktop\Chinese-LangChain-master\main.py", line 4, in
from app_modules.overwrites import postprocess
File "C:\Users\admin\Desktop\Chinese-LangChain-master\app_modules\overwrites.py", line 5, in
from app_modules.utils import *
File "C:\Users\admin\Desktop\Chinese-LangChain-master\app_modules\utils.py", line 8, in
import mdtex2html
ModuleNotFoundError: No module named 'mdtex2html'

运行5秒后报错error

can't use demo:
https://huggingface.co/spaces/ChallengeHub/Chinese-LangChain

能否就在启动知识库问答的时候，把doc下所有的文本一起加载。否则加载维基之后，本地的这些文本文件就无效了，只能再上传一个文件进行使用。

运行main.py，输入问题 提示错误，后台代码文件显示错误： AttributeError: 'Logger' object has no attribute 'warning_once'
，网上查了下没有合适的解决办法。

如题

谢谢作者的工作。
看文件说明是看corpus下中文wiki下的readme文件，但readme文件中是指数据清洗，没有提到如何生成index.faiss文件。
想请教下，如何生成index.faiss和index.pkl文件呢？（初始化后.cache文件夹下的index.pkl）

这个项目是否能实现langchain+zapier那样强大的功能呢？国内现在有没有zapier natural language api这样强大的工具呢？

Traceback (most recent call last):
File "C:\Users\86152\PycharmProjects\new\Chinese-LangChain-master\main.py", line 26, in
application = LangChainApplication(config)
File "C:\Users\86152\PycharmProjects\new\Chinese-LangChain-master\clc\langchain_application.py", line 24, in init
self.llm_service.load_model(model_name_or_path=self.config.llm_model_name)
File "C:\Users\86152\PycharmProjects\new\Chinese-LangChain-master\clc\gpt_service.py", line 55, in load_model
self.tokenizer = AutoTokenizer.from_pretrained(
File "C:\Users\86152\PycharmProjects\new\Chinese-LangChain-master\venv\lib\site-packages\transformers\models\auto\tokenization_auto.py", line 663, in from_pretrained
tokenizer_class = get_class_from_dynamic_module(
File "C:\Users\86152\PycharmProjects\new\Chinese-LangChain-master\venv\lib\site-packages\transformers\dynamic_module_utils.py", line 399, in get_class_from_dynamic_module
return get_class_in_module(class_name, final_module.replace(".py", ""))
File "C:\Users\86152\PycharmProjects\new\Chinese-LangChain-master\venv\lib\site-packages\transformers\dynamic_module_utils.py", line 177, in get_class_in_module
module = importlib.import_module(module_path)
File "D:\Ksoftware\python39\lib\importlib_init_.py", line 127, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "", line 1030, in _gcd_import
File "", line 1007, in _find_and_load
File "", line 972, in _find_and_load_unlocked
File "", line 228, in _call_with_frames_removed
File "", line 1030, in _gcd_import
File "", line 1007, in _find_and_load
File "", line 972, in _find_and_load_unlocked
File "", line 228, in _call_with_frames_removed
File "", line 1030, in _gcd_import
File "", line 1007, in _find_and_load
File "", line 984, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'transformers_modules.'
运行python main.py发生这样的错误，应该怎么解决

为啥运行app.py一直报这个错？

Hey 我是来自 langchain-serve 的dev！

请问你们在langchain部分的集成有线上的场景吗？

如果有可以考虑了解下我们的产品，方便把langchain部署在云端：

• Exposes APIs from function definitions locally as well as on the cloud.
• Very few lines of code changes, ease of development remains the same as local.
• Supports both REST & Websocket endpoints
• Serverless/autoscaling endpoints with automatic tls certs.
• Real-time streaming, human-in-the-loop support

谢谢

尊敬的Chinese-LangChain 应用开发者，我是 InternLM 社区开发者&志愿者尖米, 大佬开源的工作对我的启发很大，希望可以探讨使用 InternLM 实现Chinese-LangChain 的可能性和实现路径，我的微信是mzm312，希望可以取得联系进行更深度的交流；

环境：
Python 3.10.8
cat /proc/version
Linux version 5.4.0-99-generic (buildd@lgw01-amd64-007) (gcc version 9.3.0 (Ubuntu 9.3.0-17ubuntu120.04)) #112-Ubuntu SMP Thu Feb 3 13:50:55 UTC 2022
root@autodl-container-28d9119efa-9edf73a5:/autodl-tmp/Chinese-LangChain# cat /etc/issue
Ubuntu 22.04.1 LTS \n \l

我自己下载了模型
/root/autodl-tmp/THUDM/chatglm-6b-int4
/root/autodl-tmp/GanymedeNil/text2vec-large-chinese

做了如下修改：

llm_model_name = '/root/autodl-tmp/THUDM/chatglm-6b-int4'  # 本地模型文件 or huggingface远程仓库
embedding_model_name = '/root/autodl-tmp/GanymedeNil/text2vec-large-chinese'  # 检索模型文件 or huggingface远程仓库

运行：
python main.py
报错：

╭─────────────────────────────── Traceback (most recent call last) ────────────────────────────────╮
│ /root/autodl-tmp/Chinese-LangChain/main.py:26 in │
│ │
│ 23 │
│ 24 │
│ 25 config = LangChainCFG() │
│ ❱ 26 application = LangChainApplication(config) │
│ 27 │
│ 28 application.source_service.init_source_vector() │
│ 29 │
│ │
│ /root/autodl-tmp/Chinese-LangChain/clc/langchain_application.py:24 in init │
│ │
│ 21 │ def init(self, config): │
│ 22 │ │ self.config = config │
│ 23 │ │ self.llm_service = ChatGLMService() │
│ ❱ 24 │ │ self.llm_service.load_model(model_name_or_path=self.config.llm_model_name) │
│ 25 │ │ # self.llm_service.load_model_on_gpus(model_name_or_path=self.config.llm_model_n │
│ 26 │ │ self.source_service = SourceService(config) │
│ 27 │
│ │
│ /root/autodl-tmp/Chinese-LangChain/clc/gpt_service.py:55 in load_model │
│ │
│ 52 │ │
│ 53 │ def load_model(self, │
│ 54 │ │ │ │ model_name_or_path: str = "THUDM/chatglm-6b"): │
│ ❱ 55 │ │ self.tokenizer = AutoTokenizer.from_pretrained( │
│ 56 │ │ │ model_name_or_path, │
│ 57 │ │ │ trust_remote_code=True │
│ 58 │ │ ) │
│ │
│ /root/miniconda3/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:693 │
│ in from_pretrained │
│ │
│ 690 │ │ │ │ raise ValueError( │
│ 691 │ │ │ │ │ f"Tokenizer class {tokenizer_class_candidate} does not exist or is n │
│ 692 │ │ │ │ ) │
│ ❱ 693 │ │ │ return tokenizer_class.from_pretrained(pretrained_model_name_or_path, *input │
│ 694 │ │ │
│ 695 │ │ # Otherwise we have to be creative. │
│ 696 │ │ # if model is an encoder decoder, the encoder tokenizer class is used by default │
│ │
│ /root/miniconda3/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1812 in │
│ from_pretrained │
│ │
│ 1809 │ │ │ else: │
│ 1810 │ │ │ │ logger.info(f"loading file {file_path} from cache at {resolved_vocab_fil │
│ 1811 │ │ │
│ ❱ 1812 │ │ return cls._from_pretrained( │
│ 1813 │ │ │ resolved_vocab_files, │
│ 1814 │ │ │ pretrained_model_name_or_path, │
│ 1815 │ │ │ init_configuration, │
│ │
│ /root/miniconda3/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1975 in │
│ _from_pretrained │
│ │
│ 1972 │ │ │
│ 1973 │ │ # Instantiate tokenizer. │
│ 1974 │ │ try: │
│ ❱ 1975 │ │ │ tokenizer = cls(*init_inputs, **init_kwargs) │
│ 1976 │ │ except OSError: │
│ 1977 │ │ │ raise OSError( │
│ 1978 │ │ │ │ "Unable to load vocabulary from file. " │
│ │
│ /root/.cache/huggingface/modules/transformers_modules/chatglm-6b-int4/tokenization_chatglm.py:22 │
│ 1 in init │
│ │
│ 218 │ │ self.mask_token = mask_token │
│ 219 │ │ self.gmask_token = gmask_token │
│ 220 │ │ │
│ ❱ 221 │ │ self.sp_tokenizer = SPTokenizer(vocab_file, num_image_tokens=num_image_tokens) │
│ 222 │ │ │
│ 223 │ │ """ Initialisation """ │
│ 224 │
│ │
│ /root/.cache/huggingface/modules/transformers_modules/chatglm-6b-int4/tokenization_chatglm.py:64 │
│ in init │
│ │
│ 61 │ │ self.special_tokens = ["[MASK]", "[gMASK]", "[sMASK]", "<unused_0>", "", "< │
│ 62 │ │ self.max_blank_length = max_blank_length │
│ 63 │ │ self.byte_fallback = byte_fallback │
│ ❱ 64 │ │ self.text_tokenizer = TextTokenizer(vocab_file) │
│ 65 │ │
│ 66 │ def _get_text_tokenizer(self): │
│ 67 │ │ return self.text_tokenizer │
│ │
│ /root/.cache/huggingface/modules/transformers_modules/chatglm-6b-int4/tokenization_chatglm.py:22 │
│ in init │
│ │
│ 19 class TextTokenizer: │
│ 20 │ def init(self, model_path): │
│ 21 │ │ self.sp = spm.SentencePieceProcessor() │
│ ❱ 22 │ │ self.sp.Load(model_path) │
│ 23 │ │ self.num_tokens = self.sp.vocab_size() │
│ 24 │ │
│ 25 │ def encode(self, text): │
│ │
│ /root/miniconda3/lib/python3.10/site-packages/sentencepiece/init.py:905 in Load │
│ │
│ 902 │ │ raise RuntimeError('model_file and model_proto must be exclusive.') │
│ 903 │ if model_proto: │
│ 904 │ │ return self.LoadFromSerializedProto(model_proto) │
│ ❱ 905 │ return self.LoadFromFile(model_file) │
│ 906 │
│ 907 │
│ 908 # Register SentencePieceProcessor in _sentencepiece: │
│ │
│ /root/miniconda3/lib/python3.10/site-packages/sentencepiece/init.py:310 in LoadFromFile │
│ │
│ 307 │ │ return _sentencepiece.SentencePieceProcessor_serialized_model_proto(self) │
│ 308 │ │
│ 309 │ def LoadFromFile(self, arg): │
│ ❱ 310 │ │ return _sentencepiece.SentencePieceProcessor_LoadFromFile(self, arg) │
│ 311 │ │
│ 312 │ def _EncodeAsIds(self, text, enable_sampling, nbest_size, alpha, add_bos, add_eos, r │
│ 313 │ │ return _sentencepiece.SentencePieceProcessor__EncodeAsIds(self, text, enable_sam │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
RuntimeError: Internal: src/sentencepiece_processor.cc(1101) [model_proto->ParseFromArray(serialized.data(), serialized.size())]

是否支持apple 的M1或M2芯片？

开始上传报缺少依赖，添加后运行会崩溃。

我在main.py中新增如下接口
@app.get("/lxl")
async def root():
return {"message": "Hello World"}
采用postman请求如下

请问下，如果我想加载自己的数据库文档，txt格式。文档里面的内容应该用什么样的格式，才能使模型的效果更好呢？这方便有什么要求吗

现在的有点简略，看不明白

web搜索功能，谢谢