valuesimplex / finbert Goto Github PK

下面三个用FinBERT模型进行的实验任务的模型能够分享一下吗，想体验一下！

有没有直接进行正、负情感的情感分类？或者对应的label，从0到3表示的具体感情是？

代码：model = AutoModelForSequenceClassification.from_pretrained(checkpoint)
运行报错缺少配置文件
ValueError: Unrecognized model in XXX. Should have a model_type key in its config.json, or contain one of the following strings in its name: albert, bart, beit, bert, bert-generation, big_bird, bigbird_pegasus, blenderbot, blenderbot-small, bloom, camembert, canine, clip, codegen, convbert, convnext, ctrl, cvt, data2vec-audio, data2vec-text, data2vec-vision, deberta, deberta-v2, decision_transformer, deit, detr, distilbert, dpr, dpt, electra, encoder-decoder, flaubert, flava, fnet, fsmt, funnel, glpn, gpt2, gpt_neo, gpt_neox, gptj, groupvit, hubert, ibert, imagegpt, layoutlm, layoutlmv2, layoutlmv3, led, levit, longformer, longt5, luke, lxmert, m2m_100, marian, maskformer, mbart, mctct, megatron-bert, mobilebert, mobilevit, mpnet, mt5, mvp, nezha, nystromformer, openai-gpt, opt, owlvit, pegasus, perceiver, plbart, poolformer, prophetnet, qdqbert, rag, realm, reformer, regnet, rembert, resnet, retribert, roberta, roformer, segformer, sew, sew-d, speech-encoder-decoder, speech_to_text, speech_to_text_2, splinter, squeezebert, swin, t5, tapas, trajectory_transformer, transfo-xl, trocr, unispeech, unispeech-sat, van, vilt, vision-encoder-decoder, vision-text-dual-encoder, visual_bert, vit, vit_mae, wav2vec2, wav2vec2-conformer, wavlm, xglm, xlm, xlm-prophetnet, xlm-roberta, xlm-roberta-xl, xlnet, yolos, yoso

希望能开放点数据集，万分感谢，邮箱[email protected]

bert还是太重了，希望后续有albert，albert_tiny等版本！对于训练数据，虽然不切实际的幻想开放，万一开放呢？

model = BertForSequenceClassification.from_pretrained('/content/sample_data/FinBERT_pytorch/bert_config.json',from_tf = True)
以下是报错：

AttributeError Traceback (most recent call last)
in ()
----> 1 model = BertForSequenceClassification.from_pretrained('/content/sample_data/FinBERT_pytorch/bert_config.json',from_tf = True)

/usr/local/lib/python3.6/dist-packages/transformers/modeling_utils.py in from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs)
970 from .modeling_tf_pytorch_utils import load_tf2_checkpoint_in_pytorch_model
971
--> 972 model = load_tf2_checkpoint_in_pytorch_model(model, resolved_archive_file, allow_missing_keys=True)
973 except ImportError:
974 logger.error(

/usr/local/lib/python3.6/dist-packages/transformers/modeling_tf_pytorch_utils.py in load_tf2_checkpoint_in_pytorch_model(pt_model, tf_checkpoint_path, tf_inputs, allow_missing_keys)
266 import transformers
267
--> 268 from .modeling_tf_utils import load_tf_weights
269
270 logger.info("Loading TensorFlow weights from {}".format(tf_checkpoint_path))

/usr/local/lib/python3.6/dist-packages/transformers/modeling_tf_utils.py in ()
1029
1030
-> 1031 def get_initializer(initializer_range: float = 0.02) -> tf.initializers.TruncatedNormal:
1032 """
1033 Creates a :obj:tf.initializers.TruncatedNormal with the given range.

/usr/local/lib/python3.6/dist-packages/tensorflow/python/util/deprecation_wrapper.py in getattr(self, name)
104 if name.startswith('dw'):
105 raise AttributeError('Accessing local variables before they are created.')
--> 106 attr = getattr(self._dw_wrapped_module, name)
107 if (self._dw_warning_count < _PER_MODULE_WARNING_LIMIT and
108 name not in self._dw_deprecated_printed):

AttributeError: module 'tensorflow._api.v1.initializers' has no attribute 'TruncatedNormal'

请问依赖的 transformers 的版本是什么？
另外，tokenizer = BertTokenizer.from_pretrained('/content/sample_data/FinBERT_pytorch',from_tf=True)，这行命令是可以运行的

1.0版本效果很棒，比wiki中文训练的bert效果好了几个百分点，期待后续版本的开源。

代码：
model = BertForSequenceClassification.from_pretrained('/home/linux/FinBERT/bert_config.json',from_tf=True)
加载报错 ，安装tensorflow==1.14.0 会报错data_adapter 和 .keras_tensor 模块找不到

~/anaconda3/lib/python3.6/site-packages/transformers/modeling_tf_utils.py in <module>() 29 from tensorflow.python.keras import backend as K 30 from tensorflow.python.keras.engine import data_adapter ---> 31 from tensorflow.python.keras.engine.keras_tensor import KerasTensor 32 from tensorflow.python.keras.saving import hdf5_format 33 ModuleNotFoundError: No module named 'tensorflow.python.keras.engine.keras_tensor'

能用该模型对股票投资视频的英文字幕进行翻译啊？常规的翻译模型翻译结果太差了，你们的模型对翻译质量有显著提升吗？

代码:
model = BertForSequenceClassification.from_pretrained(model_name+"/bert_config.json") model.to(device)
报错：
OSError: Unable to load weights from pytorch checkpoint file. If you tried to load a PyTorch model from a TF 2.0 checkpoint, please set from_tf=True.

现有的finbert是基于base版bert模型训练，那么后续的large版本有计划开源？