Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / CyberZHG/keras-bert issues and pull requests

#226 - 'adamwarmup' object has no attribute '_set_hyper'

Issue - State: open - Opened by jianlianggao about 1 year ago
Labels: bug

#221 - module 'tensorflow.compat.v2' has no attribute '__internal__'

Issue - State: closed - Opened by TerryTT2 over 3 years ago - 1 comment
Labels: question, wontfix

#220 - AttributeError: module 'keras' has no attribute 'applications'

Issue - State: closed - Opened by vngrv over 3 years ago - 5 comments
Labels: question, wontfix

#219 - Masked LM pre-training

Issue - State: closed - Opened by Khedesh over 3 years ago - 1 comment
Labels: question, wontfix

#218 - Error in getting attention map

Issue - State: closed - Opened by suyashh94 over 3 years ago - 3 comments
Labels: bug, wontfix

#217 - Improve Tokenizer for uppercase text

Issue - State: closed - Opened by HariWu1995 over 3 years ago - 1 comment
Labels: question, wontfix

#216 - 关于显存巨量占用的疑惑

Issue - State: closed - Opened by Creling over 3 years ago - 1 comment
Labels: question, wontfix

#215 - which version of keras-bert is compatible for tensorflow 1.13?

Issue - State: closed - Opened by vikotse over 3 years ago - 2 comments
Labels: question

#214 - API reference list

Issue - State: closed - Opened by UCRajkumar over 3 years ago
Labels: question

#213 - The method get_model to get the bert model is right?

Issue - State: closed - Opened by lwyeah almost 4 years ago - 4 comments
Labels: bug

#212 - Keras Bert with character embeddings

Issue - State: closed - Opened by shenron9155 almost 4 years ago - 1 comment
Labels: question, wontfix

#211 - 在哪里可以找到之前的版本呢？

Issue - State: closed - Opened by Ignoranceemperor about 4 years ago - 2 comments
Labels: question, wontfix

#210 - List index out of range for example code

Issue - State: closed - Opened by tama-biro about 4 years ago - 2 comments
Labels: question, wontfix

#209 - Input sentence for extract_embeddings

Issue - State: closed - Opened by bigredbug47 about 4 years ago - 1 comment
Labels: question, wontfix

#208 - 如何提取长度一致的bert词向量？

Issue - State: closed - Opened by 1535966643 about 4 years ago - 1 comment
Labels: question, wontfix

#207 - 如何实现在自己的数据上很少的预训练？

Issue - State: closed - Opened by 1535966643 about 4 years ago - 1 comment
Labels: question, wontfix

#206 - could you provide an albert loader ?

Issue - State: closed - Opened by ghost about 4 years ago - 1 comment
Labels: question, wontfix

#205 - keras-bert load pre model failed in keras 2.2.4, maybe it is a bug

Issue - State: closed - Opened by ghost about 4 years ago
Labels: bug

#204 - 如何提取fine tune之后模型的字（词）向量？

Issue - State: closed - Opened by JackiyWu about 4 years ago - 1 comment
Labels: question, wontfix

#203 - how can do ner task in keras-bert ? anyone has applied the crf in keras-bert?

Issue - State: closed - Opened by ghost about 4 years ago
Labels: question

#202 - How to transfer character vector into word vector for Chinese texts

Issue - State: closed - Opened by JackiyWu about 4 years ago - 1 comment
Labels: question, wontfix

#201 - How to use bert encoder for continuous values directly? [ No pre-trained model ]

Issue - State: closed - Opened by monk1337 about 4 years ago - 1 comment
Labels: question, wontfix

#200 - keras-bert怎么使用多个gpu进行训练？

Issue - State: closed - Opened by jane96 about 4 years ago - 1 comment
Labels: question, wontfix

#199 - Is this model available to be converted to a .tflite?

Issue - State: closed - Opened by hguan8 over 4 years ago - 1 comment
Labels: question, wontfix

#198 - compatibility with tf 2.3

Pull Request - State: closed - Opened by tim5go over 4 years ago - 2 comments

#197 - How to save and load Python Keras BERT model to serialize it?

Issue - State: closed - Opened by lxykyle01 over 4 years ago - 1 comment
Labels: question

#196 - Which version of keras-bert should I download? keras:2.2.4 tf:1.12.0

Issue - State: closed - Opened by AliveHH over 4 years ago - 1 comment
Labels: question

#195 - NoneType error when fine tuning BERT CLS embeddings directly

Issue - State: closed - Opened by radurevutchi over 4 years ago - 2 comments
Labels: question, wontfix

#194 - error on codec call, no file found

Issue - State: closed - Opened by troy109 over 4 years ago - 2 comments
Labels: bug

#192 - Position embedding for relation extraction

Issue - State: closed - Opened by juanluis17 over 4 years ago - 1 comment
Labels: question, wontfix

#191 - Removed "adapter"

Issue - State: closed - Opened by CyberZHG over 4 years ago - 1 comment

#190 - keras-bert relies on features removed in keras-transformer 0.34.0

Issue - State: closed - Opened by joseph-wakeling-frequenz over 4 years ago - 7 comments
Labels: bug, wontfix

#189 - Training a model from scratch?

Issue - State: closed - Opened by ddofer over 4 years ago - 1 comment
Labels: question, wontfix

#188 - The loss of MLM seems wrong

Issue - State: closed - Opened by showerage over 4 years ago
Labels: bug

#187 - WARNING:tensorflow:Entity

Issue - State: closed - Opened by kulturevulture over 4 years ago - 3 comments
Labels: bug, wontfix

#186 - Attention maps

Issue - State: closed - Opened by jogonba2 over 4 years ago - 1 comment
Labels: question

#185 - fine-tuning without using NSP-Dense

Issue - State: closed - Opened by kuroshiba-Ginji over 4 years ago - 1 comment
Labels: wontfix

#184 - Concatenate BERT and CNN

Issue - State: closed - Opened by efabb95 over 4 years ago - 7 comments
Labels: bug, wontfix

#183 - TypeError in gelu when using different float precision

Issue - State: closed - Opened by f-kretschmer almost 5 years ago - 1 comment
Labels: bug, wontfix

#181 - Fix compatibility

Pull Request - State: closed - Opened by CyberZHG almost 5 years ago

#180 - How to prepare the input of test data?

Issue - State: closed - Opened by houzeyu2683 almost 5 years ago - 1 comment
Labels: question, wontfix

#179 - Question about the ordering of LayerNorm and Dropout.

Issue - State: closed - Opened by Alkaiddd almost 5 years ago - 1 comment
Labels: question, wontfix

#178 - Slow inference on TPU with Colaboratory

Issue - State: closed - Opened by alex4321 almost 5 years ago - 1 comment
Labels: wontfix

#177 - error happens in the next-sentence prediction case

Issue - State: closed - Opened by TiranoGreatLand almost 5 years ago - 1 comment
Labels: bug, wontfix

#176 - Add words to a pretrained model

Issue - State: closed - Opened by dberma15 almost 5 years ago - 1 comment
Labels: question, wontfix

#175 - POOL_NSP

Issue - State: closed - Opened by zhangxin9988 almost 5 years ago - 1 comment
Labels: question, wontfix

#174 - 请问这个包对tensorflow的版本没有要求吗

Issue - State: closed - Opened by duterscmy almost 5 years ago - 1 comment
Labels: question, wontfix

#173 - fine-tuning

Issue - State: closed - Opened by OKtang almost 5 years ago - 3 comments
Labels: question, wontfix

#172 - 关于模型的加载和修改

Issue - State: closed - Opened by yuhao1982 almost 5 years ago - 1 comment
Labels: question, wontfix

#171 - Attention mask是不需要吗？

Issue - State: closed - Opened by BaberMuyu almost 5 years ago - 5 comments
Labels: question

#170 - Run extract_embeddings with GPU

Issue - State: closed - Opened by 336655asd about 5 years ago - 3 comments
Labels: question, wontfix

#169 - failed to create cublas handle: CUBLAS_STATUS_NOT_INITIALIZED

Issue - State: closed - Opened by Imorton-zd about 5 years ago
Labels: bug

#168 - support albert?

Issue - State: closed - Opened by ghost about 5 years ago - 1 comment
Labels: enhancement, wontfix

#167 - AdamWarmup not working in transformers

Issue - State: closed - Opened by acmilannesta about 5 years ago - 3 comments
Labels: question, wontfix

#166 - How can I input padding mask?

Issue - State: closed - Opened by xinbingzhe about 5 years ago - 2 comments
Labels: question, wontfix

#165 - 怎么使用tensorflow的keras

Issue - State: closed - Opened by lx-rookie about 5 years ago - 1 comment
Labels: question, wontfix

#164 - Is there a way to convert a tf.keras saved model to keras?

Issue - State: closed - Opened by gargomeiste about 5 years ago - 1 comment
Labels: enhancement, wontfix

#163 - I would like to embed Bert Embeddings Layer and use it in my model for Word Sense Disambiguation

Issue - State: closed - Opened by elsheikh21 about 5 years ago - 1 comment
Labels: question, wontfix

#162 - AttributeError: 'tuple' object has no attribute 'layer'

Issue - State: closed - Opened by linWujl about 5 years ago - 4 comments
Labels: question, wontfix

#161 - suport for conda？

Issue - State: closed - Opened by lidadreamer about 5 years ago - 1 comment
Labels: question, wontfix

#160 - checkpoints being auto deleted

Issue - State: closed - Opened by XIE-Fangyuan about 5 years ago - 1 comment
Labels: question, wontfix

#159 - Albert support

Issue - State: closed - Opened by flydsc about 5 years ago - 2 comments
Labels: enhancement, wontfix

#158 - Further question about trainable and trainning

Issue - State: closed - Opened by weizhenzhao about 5 years ago - 4 comments
Labels: question, wontfix

#157 - GPU memory use too much after load trained model

Issue - State: closed - Opened by moriwang about 5 years ago - 1 comment
Labels: question, wontfix

#156 - When I turn the batch_size up or change to the bert_large model, I get the following error.

Issue - State: closed - Opened by George191 about 5 years ago - 2 comments
Labels: question, wontfix

#155 - support 3-d input

Issue - State: closed - Opened by lsq357 about 5 years ago - 1 comment
Labels: question, wontfix

#154 - Warning Message after updates ver0.80.0.0

Issue - State: closed - Opened by Xiaoping777 about 5 years ago - 1 comment
Labels: bug, wontfix

#153 - support albert?

Issue - State: closed - Opened by yongzhuo about 5 years ago - 1 comment
Labels: enhancement, wontfix

#152 - How to save fine tuned bert model with origin bert format

Issue - State: closed - Opened by luzhongqiu about 5 years ago - 1 comment
Labels: question, wontfix

#151 - I get some problem when I do the task of cws with keras-bert

Issue - State: closed - Opened by kiko441500 about 5 years ago - 1 comment
Labels: question, wontfix

#150 - 关于如何使用训练的模型

Issue - State: closed - Opened by yanqiangmiffy about 5 years ago
Labels: question

#149 - How to use FP16 mode?

Issue - State: closed - Opened by ghost about 5 years ago - 1 comment
Labels: question, wontfix

#148 - I want to add a few special words to Pre-training-bert.

Issue - State: closed - Opened by ghost about 5 years ago - 1 comment
Labels: question

#147 - Slow predictions

Issue - State: closed - Opened by richfolsom about 5 years ago - 1 comment
Labels: bug, wontfix

#146 - MaskMaxPooling1D

Issue - State: closed - Opened by rortegagit about 5 years ago - 1 comment
Labels: enhancement, wontfix

#145 - bert_model.fit_generator不能保存模型

Issue - State: closed - Opened by yanqiangmiffy about 5 years ago - 1 comment
Labels: invalid

#144 - bert model could not applied in TimeDistributed wrapper.

Issue - State: closed - Opened by phoenixkillerli about 5 years ago - 1 comment
Labels: invalid

#143 - get_updates() takes 3 positional arguments but 4 were given

Issue - State: closed - Opened by pengchenghu428 about 5 years ago - 1 comment
Labels: wontfix

#142 - keras.utils.multi_gpu_model() does not work

Issue - State: closed - Opened by superxuxin about 5 years ago - 3 comments
Labels: invalid, wontfix

#141 - Load model error, .h5 file load error: error message = 'Is a directory'

Issue - State: closed - Opened by luzhongqiu about 5 years ago - 4 comments
Labels: invalid

#140 - model.predict([np.array([indices]), np.array([segments]), masks])[1]

Issue - State: closed - Opened by LyJss about 5 years ago - 1 comment
Labels: question

#139 - Remove tensorflow warnings

Issue - State: closed - Opened by BrikerMan about 5 years ago - 3 comments
Labels: bug, wontfix

#138 - Tokenization

Issue - State: closed - Opened by sonfire186 about 5 years ago - 1 comment
Labels: question, wontfix

#137 - Question about why transpose Weight Matrix when load "NSP" layer (THX)

Issue - State: closed - Opened by Zenghao-CQ about 5 years ago - 2 comments
Labels: question

#136 - Pathological memory usage in training, scaling with number of examples

Issue - State: closed - Opened by LordZorgoth about 5 years ago - 7 comments
Labels: bug, wontfix

#135 - Use the trained model?

Issue - State: closed - Opened by statikkkkk about 5 years ago - 4 comments
Labels: question, wontfix

#134 - How can I get the output of the last few layers?

Issue - State: closed - Opened by Chevalier1024 about 5 years ago
Labels: question

#133 - Masked Language modelling only predicts padding tokens

Issue - State: closed - Opened by RAXAI about 5 years ago - 1 comment
Labels: question, wontfix

#132 - How do I load your model's predictions as inputs to my model via fit_generator?

Issue - State: closed - Opened by moonman239 about 5 years ago - 2 comments
Labels: question, wontfix

#131 - Self-Attention Mask

Issue - State: closed - Opened by Azharo about 5 years ago - 1 comment
Labels: question

#128 - 想精简模型，把bert的层数降低，再人工预训练模型，有相应的Demo吗

Issue - State: closed - Opened by li1191863273 about 5 years ago - 1 comment
Labels: question, wontfix

#127 - MLM训练的机制

Issue - State: closed - Opened by dengkunn about 5 years ago - 15 comments
Labels: question, wontfix

#116 - TypeError: get_encoders() got an unexpected keyword argument 'use_adapter'

Issue - State: closed - Opened by MNCTTY over 5 years ago - 11 comments
Labels: invalid

#114 - TypeError: get_encoders() got an unexpected keyword argument 'use_adapter'

Issue - State: closed - Opened by nwth over 5 years ago - 4 comments
Labels: question

#100 - Error while predicting using a saved model

Issue - State: closed - Opened by devinharia over 5 years ago - 5 comments
Labels: question, wontfix

#78 - ValueError: Expected to see 3 array(s)

Issue - State: closed - Opened by xinsu626 over 5 years ago - 2 comments