Tongjilibo/bert4torch issues and pull requests

#74 - TSDAE测试结果异常，spearman分值极低甚至会NAN，检查dataset和collate_fn无问题，请问可能是什么原因？

Issue - State: open - Opened by rutonggaoya about 2 years ago - 1 comment
Labels: help wanted

#73 - task_sequence_labeling_ner_tplinker_plus报错，f1为0

Issue - State: closed - Opened by LiangXD about 2 years ago - 12 comments

#72 - 想调用未内置的模型的请进

Issue - State: open - Opened by Tongjilibo about 2 years ago

#71 - 请问如何用roformer-sim来实现相似句二分类任务啊

Issue - State: closed - Opened by aph-asic about 2 years ago - 8 comments

#70 - 分类模型转onnx的时候报如下错误

Issue - State: closed - Opened by wanglaiqi about 2 years ago - 1 comment

#69 - 问题

Issue - State: closed - Opened by lili-li-cpu about 2 years ago - 10 comments

#68 - MRC式抽取模型与原文是不是有点出入

Issue - State: open - Opened by Bureaux-Tao about 2 years ago - 1 comment

#67 - 对further pretrain的训练数据的格式要求？

Issue - State: closed - Opened by XuJianzhi about 2 years ago - 5 comments
Labels: duplicate

#66 - 大佬，有意复现这篇CNER吗？

Issue - State: open - Opened by nlper01 about 2 years ago - 3 comments
Labels: todo list

#65 - 请问UIE模型的bin文件在哪里下载的

Issue - State: closed - Opened by YLiberals about 2 years ago - 2 comments

#64 - 关于bert.is_dropout参数的理解

Issue - State: closed - Opened by XuJianzhi about 2 years ago - 4 comments

#63 - 大佬您好，我想问一下roberta的mlm预训练的数据长什么样

Issue - State: closed - Opened by swx-10 over 2 years ago - 11 comments

#62 - 多机多卡分布式训练如何实现？

Issue - State: open - Opened by 447428054 over 2 years ago - 4 comments
Labels: todo list

#61 - 关于xlnet的分词器问题

Issue - State: closed - Opened by QAQ112 over 2 years ago - 6 comments

#60 - task_sequence_labeling_ner_CNN_Nested_NER运行问题

Issue - State: closed - Opened by QAQ112 over 2 years ago - 3 comments

#59 - metrics问题

Issue - State: closed - Opened by Alwin4Zhang over 2 years ago - 2 comments

#58 - 如何在bert4torch下使用tensorboard ？

Issue - State: closed - Opened by XuJianzhi over 2 years ago - 8 comments

#57 - 关系抽取

Issue - State: closed - Opened by xx-Jiangwen over 2 years ago - 6 comments

#56 - 用于英文数据应该改哪些？

Issue - State: closed - Opened by XuJianzhi over 2 years ago - 1 comment

#55 - EfficientGlobalPointer实现问题

Issue - State: closed - Opened by Alwin4Zhang over 2 years ago - 1 comment

#54 - 您好，不能开启多线程

Issue - State: closed - Opened by fuxuelinwudi over 2 years ago - 14 comments

#53 - 作者您好，请问您以后会更新关系抽取例子中的代码对测试集的验证吗，我自己尝试去加，但能力不够

Issue - State: closed - Opened by Liekkas22years over 2 years ago - 5 comments

#52 - 一个问题

Issue - State: closed - Opened by lili-li-cpu over 2 years ago - 7 comments

#51 - 模型结果不是特别好，不知道是什么问题

Issue - State: closed - Opened by nlper01 over 2 years ago - 7 comments

#50 - 关于gradient-checkpointing的支持

Issue - State: closed - Opened by boxiaowave over 2 years ago - 3 comments

#49 - 大佬您好，看了您写的代码感觉非常棒，请问您后续打算用bert4torch复现CNN_Nested_NER么

Issue - State: closed - Opened by 1365717822 over 2 years ago - 4 comments

#48 - csl数据集文本摘要

Issue - State: closed - Opened by fuhaito7 over 2 years ago - 2 comments

#47 - 预训练模型加载出错的请进

Issue - State: open - Opened by Tongjilibo over 2 years ago - 2 comments
Labels: documentation

#46 - seq2seq_bart模型出错

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

#45 - 一个小问题

Issue - State: closed - Opened by lili-li-cpu over 2 years ago - 2 comments

#42 - 虚拟对抗训练

Issue - State: closed - Opened by lili-li-cpu over 2 years ago - 15 comments

#41 - 您好，请问W2NER的预测怎么写呢

Issue - State: closed - Opened by zyx214 over 2 years ago - 3 comments

#40 - tensorrt 把onnx转化成fp16的trt engine时，精度损失

Issue - State: open - Opened by chironma over 2 years ago - 1 comment

#37 - 您好，波哥，我想在关系抽取模型上添加对抗训练方法，预训练模型使用的是ernie模型，请问需要改预训练模型的embding层嘛

Issue - State: closed - Opened by lili-li-cpu over 2 years ago - 9 comments

#31 - 模型训练中的收敛问题

Issue - State: closed - Opened by Bobo1204 over 2 years ago - 5 comments

#25 - 使用task_seq2seq_autotitle_csl_uer_t5.py代码，加载uer/t5-small-chinese-cluecorpussmall中的config.json报错

Issue - State: closed - Opened by semon2019 over 2 years ago - 6 comments
Labels: documentation

#24 - Traceback (most recent call last): File "task_sentiment_classification_hierarchical_position.py", line 117, in <module> model.fit(train_dataloader, epochs=10, steps_per_epoch=None, callbacks=[evaluator]) File "/home/huangjiaheng/.local/lib/python3.8/site-packages/bert4torch/models.py", line 274, in fit loss.backward(retain_graph=retain_graph) File "/opt/conda/lib/python3.8/site-packages/torch/tensor.py", line 245, in backward torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) File "/opt/conda/lib/python3.8/site-packages/torch/autograd/init.py", line 145, in backward Variable._execution_engine.run_backward( RuntimeError: cuda runtime error (710) : device-side assert triggered at /opt/conda/conda-bld/pytorch_1616554793803/work/aten/src/THC/generic/THCTensorMath.cu:29

Issue - State: closed - Opened by huangjiaheng over 2 years ago - 1 comment

#23 - 运行W2NER示例是报错ValueError: too many values to unpack (expected 2)

Issue - State: closed - Opened by nlper01 over 2 years ago - 8 comments

#18 - basic_language_model_nezha_gen_gpt.py

Issue - State: closed - Opened by chongzicbo over 2 years ago - 2 comments
Labels: documentation

#17 - 遇到的几个问题分享一下，以及博主波哥的帮助解决记录

Issue - State: closed - Opened by bestgwl over 2 years ago

#15 - basic_masked_language_model.py

Issue - State: closed - Opened by chongzicbo over 2 years ago - 1 comment
Labels: duplicate

#13 - bert+crf 做ner任务的基础案例出现loss 陡增情况

Issue - State: closed - Opened by zbbwss over 2 years ago - 5 comments
Labels: bug

#10 - RuntimeError: masked_select: expected BoolTensor or ByteTensor for mask

Issue - State: closed - Opened by zbbwss over 2 years ago - 8 comments

#9 - 一个可能的小问题

Issue - State: closed - Opened by Uchiha-Monroe over 2 years ago - 1 comment

GitHub / Tongjilibo/bert4torch issues and pull requests