运行对话脚本aquilachat2-34b-16k 报错误
gaord opened this issue · comments
System Info
OS macos
flagai 1.8.2
transformers 4.34.0
错误如下:
[2023-10-13 13:25:52,131] [INFO] [logger.py:85:log_dist] [Rank -1] Unsupported bmtrain
Loading checkpoint shards: 100%|██████████████████████████████████████████████████████████████| 7/7 [00:16<00:00, 2.31s/it]
The repository for ./checkpoints/aquilachat2-34b-16k contains custom code which must be executed to correctlyload the model. You can inspect the repository content at https://hf.co/./checkpoints/aquilachat2-34b-16k.
You can avoid this prompt in future by passing the argument trust_remote_code=True
.
Do you wish to run the custom code? [y/N] y
Traceback (most recent call last):
File "/Users/pom/AIGC/run-Aquila.py", line 11, in
autoloader = AutoLoader("aquila2", model_name=model_name, device=device)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/homebrew/lib/python3.11/site-packages/flagai/auto_model/auto_loader.py", line 305, in init
tokenizer = AutoTokenizer.from_pretrained(download_path)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/homebrew/lib/python3.11/site-packages/transformers/models/auto/tokenization_auto.py", line 779, in from_pretrained
raise ValueError(
ValueError: Unrecognized configuration class <class 'transformers_modules.aquilachat2-34b-16k.configuration_aquila.AquilaConfig'> to build an AutoTokenizer.
Model type should be one of AlbertConfig, AlignConfig, BarkConfig, BartConfig, BertConfig, BertGenerationConfig, BigBirdConfig, BigBirdPegasusConfig, BioGptConfig, BlenderbotConfig, BlenderbotSmallConfig, BlipConfig, Blip2Config, BloomConfig, BridgeTowerConfig, BrosConfig, CamembertConfig, CanineConfig, ChineseCLIPConfig, ClapConfig, CLIPConfig, CLIPSegConfig, LlamaConfig, CodeGenConfig, ConvBertConfig, CpmAntConfig, CTRLConfig, Data2VecAudioConfig, 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, IdeficsConfig, InstructBlipConfig, JukeboxConfig, LayoutLMConfig, LayoutLMv2Config, LayoutLMv3Config, LEDConfig, LiltConfig, LlamaConfig, LongformerConfig, LongT5Config, LukeConfig, LxmertConfig, M2M100Config, MarianConfig, MBartConfig, MegaConfig, MegatronBertConfig, MgpstrConfig, MistralConfig, MobileBertConfig, MPNetConfig, MptConfig, MraConfig, MT5Config, MusicgenConfig, MvpConfig, NezhaConfig, NllbMoeConfig, NystromformerConfig, OneFormerConfig, OpenAIGPTConfig, OPTConfig, OwlViTConfig, PegasusConfig, PegasusXConfig, PerceiverConfig, PersimmonConfig, Pix2StructConfig, PLBartConfig, ProphetNetConfig, QDQBertConfig, RagConfig, RealmConfig, ReformerConfig, RemBertConfig, RetriBertConfig, RobertaConfig, RobertaPreLayerNormConfig, RoCBertConfig, RoFormerConfig, RwkvConfig, Speech2TextConfig, Speech2Text2Config, SpeechT5Config, SplinterConfig, SqueezeBertConfig, SwitchTransformersConfig, T5Config, TapasConfig, TransfoXLConfig, UMT5Config, ViltConfig, VisualBertConfig, VitsConfig, Wav2Vec2Config, Wav2Vec2ConformerConfig, WhisperConfig, XCLIPConfig, XGLMConfig, XLMConfig, XLMProphetNetConfig, XLMRobertaConfig, XLMRobertaXLConfig, XLNetConfig, XmodConfig, YosoConfig.
Information
- The official example scripts
- My own modified scripts
Tasks
- An officially supported task in the
examples
folder (such as T5/AltCLIP, ...) - My own task or dataset (give details below)
Reproduction
按照网页提供的步骤准备环境
Expected behavior
能够运行官方提供的推断脚本
之前我们的bug,模型tokenizer文件不完整。您更新试试。