Transformer_models-源码
BERT在Azure机器学习服务上
此回购包含终端到终端的食谱和的 （双向编码器交涉来自变形金刚）用语言表达模型。
伯特
BERT是一种语言表示模型，其特征在于可以有效捕获语料库中深层和微妙的文本关系。 在原始论文中，作者证明了BERT模型可以很容易地改编以构建用于许多NLP任务的最新模型，包括文本分类，命名实体识别和问题解答。 在此仓库中，我们提供了笔记本，使开发人员可以从语料库中重新训练BERT模型，并微调现有的BERT模型以解决专门的任务。 此回购中提供了的简要可快速开始使用BERT。
预训练
BERT预训练中的挑战
将BERT语言表示模型预训练到所需的准确性水平是非常具有挑战性的。 结果，大多数开发人员从在标准语料库（例如Wikipedia）上经过预训练的BERT模型开始，而不是从头开始训练它。 如果在与训练前步骤中使用的语料库相似的语料库上训练最终模型，则此策略效果很好。 但是，
文件下载
资源详情
[{"title":"（ 52 个子文件 232KB ） Transformer_models-源码","children":[{"title":"transformer_models-master","children":[{"title":"docs","children":[{"title":"bert-intro.md <span style='color:#111;'> 5.01KB </span>","children":null,"spread":false},{"title":"artifacts.md <span style='color:#111;'> 2.63KB </span>","children":null,"spread":false},{"title":"dataprep.md <span style='color:#111;'> 1.13KB </span>","children":null,"spread":false}],"spread":true},{"title":"huggingface_models","children":[{"title":"modeling_electra.py <span style='color:#111;'> 21.83KB </span>","children":null,"spread":false}],"spread":true},{"title":"pretrain","children":[{"title":"PyTorch","children":[{"title":"models.py <span style='color:#111;'> 10.96KB </span>","children":null,"spread":false},{"title":"text.py <span style='color:#111;'> 107B </span>","children":null,"spread":false},{"title":"utils.py <span style='color:#111;'> 356B </span>","children":null,"spread":false},{"title":"azureml_adapter.py <span style='color:#111;'> 1.48KB </span>","children":null,"spread":false},{"title":"logger.py <span style='color:#111;'> 697B </span>","children":null,"spread":false},{"title":"train.py <span style='color:#111;'> 24.15KB </span>","children":null,"spread":false},{"title":"dataset.py <span style='color:#111;'> 6.72KB </span>","children":null,"spread":false},{"title":"benchmark.py <span style='color:#111;'> 1.24KB </span>","children":null,"spread":false},{"title":"configuration.py <span style='color:#111;'> 1.71KB </span>","children":null,"spread":false},{"title":"dataprep","children":[{"title":"create_pretraining.py <span style='color:#111;'> 4.01KB </span>","children":null,"spread":false},{"title":"single_line_doc_file_creation.py <span style='color:#111;'> 905B </span>","children":null,"spread":false},{"title":"split_data_into_files.py <span style='color:#111;'> 1.54KB </span>","children":null,"spread":false},{"title":"sentence_segmentation.py <span style='color:#111;'> 612B </span>","children":null,"spread":false}],"spread":false},{"title":"distributed_apex.py <span style='color:#111;'> 24.49KB </span>","children":null,"spread":false},{"title":"sources.py <span style='color:#111;'> 9.34KB </span>","children":null,"spread":false},{"title":"checkpoint.py <span style='color:#111;'> 2.45KB </span>","children":null,"spread":false},{"title":"optimization.py <span style='color:#111;'> 1.17KB </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 341B </span>","children":null,"spread":false},{"title":"notebooks","children":[{"title":"BERT_Pretrain.ipynb <span style='color:#111;'> 17.14KB </span>","children":null,"spread":false}],"spread":false}],"spread":false},{"title":"configs","children":[{"title":"bert-large-single-node.json <span style='color:#111;'> 1019B </span>","children":null,"spread":false},{"title":"bert-base-single-node-rui.json <span style='color:#111;'> 1.04KB </span>","children":null,"spread":false},{"title":"bert-base.json <span style='color:#111;'> 987B </span>","children":null,"spread":false},{"title":"electra-small-single-node-rui.json <span style='color:#111;'> 1.29KB </span>","children":null,"spread":false},{"title":"bert-base-single-node.json <span style='color:#111;'> 1013B </span>","children":null,"spread":false},{"title":"bert-large.json <span style='color:#111;'> 993B </span>","children":null,"spread":false}],"spread":true},{"title":"README.md <span style='color:#111;'> 2.34KB </span>","children":null,"spread":false},{"title":"scripts","children":[{"title":"single_node_bert.lsf <span style='color:#111;'> 1.46KB </span>","children":null,"spread":false},{"title":"single_node_electra.lsf <span style='color:#111;'> 1.53KB </span>","children":null,"spread":false}],"spread":true}],"spread":true},{"title":"finetune","children":[{"title":"run_classifier_azureml.py <span style='color:#111;'> 45.75KB </span>","children":null,"spread":false},{"title":"evaluate_squad.py <span style='color:#111;'> 3.34KB </span>","children":null,"spread":false},{"title":"PyTorch","children":[{"title":"run_finetuning_glue.py <span style='color:#111;'> 37.09KB </span>","children":null,"spread":false},{"title":"models.py <span style='color:#111;'> 4.96KB </span>","children":null,"spread":false},{"title":"azureml_bert_util.py <span style='color:#111;'> 4.65KB </span>","children":null,"spread":false},{"title":"notebooks","children":[{"title":"Pretrained-BERT-GLUE.ipynb <span style='color:#111;'> 19.20KB </span>","children":null,"spread":false},{"title":"BERT_Eval_GLUE.ipynb <span style='color:#111;'> 13.44KB </span>","children":null,"spread":false},{"title":"Pretrained-BERT-NER.ipynb <span style='color:#111;'> 46.99KB </span>","children":null,"spread":false},{"title":"BERT_Eval_SQUAD.ipynb <span style='color:#111;'> 12.66KB </span>","children":null,"spread":false}],"spread":true},{"title":"run_finetuning_glue_horovodrun.py <span style='color:#111;'> 34.22KB </span>","children":null,"spread":false},{"title":"dockerfile <span style='color:#111;'> 317B </span>","children":null,"spread":false}],"spread":true},{"title":"run_squad_azureml.py <span style='color:#111;'> 45.84KB </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 9.75KB </span>","children":null,"spread":false},{"title":"scripts","children":[{"title":"single_node_bert_fine_tuning.lsf <span style='color:#111;'> 2.51KB </span>","children":null,"spread":false},{"title":"single_node_electra_fine_tuning.lsf <span style='color:#111;'> 2.55KB </span>","children":null,"spread":false}],"spread":true},{"title":"TensorFlow","children":[{"title":"run_classifier.py <span style='color:#111;'> 32.15KB </span>","children":null,"spread":false},{"title":"download_model_and_dataset.py <span style='color:#111;'> 3.04KB </span>","children":null,"spread":false},{"title":"notebooks","children":[{"title":"Tensorflow-BERT-AzureML.ipynb <span style='color:#111;'> 139.93KB </span>","children":null,"spread":false}],"spread":false}],"spread":true}],"spread":true},{"title":"LICENSE <span style='color:#111;'> 1.13KB </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 6.31KB </span>","children":null,"spread":false}],"spread":true}],"spread":true}]
评论信息
其他资源
- python 京东预约抢购茅台脚本插件 一键运行
- 西安市完整矢量shp数据
- 60KW充电桩电路图
- 非线性控制时滞系统迭代学习控制算法仿真
- 网上书店系统
- python脚本运行环境.zip
- QC11.0安装包
- 嵌入式硬件设计(中文版)(老外的经典好书)pdf
- Microwind2和DSCH2
- 基于矢量量化的0-9孤立词语音识别
- 动态心电图分析软件支持TLC3000 TLC4000 直接使用。
- ios仿购物系统源代码
- MATLAB\心电滤波效果查看器GUI设计
- 2019最新邮政编码大全.sql
- 基于docker部署前后端分离项目文档
- 图解CSS3核心技术与案例实战.pdf
- 论文撰写辅助工具.jar
- transmission-web-control-1.6.0-beta2
- 模糊控制及其MATLAB仿真
- Pearson 电路第十版英文非扫描版
- Linux命令行与shell脚本编程大全.第3版【高清 带书签】
- RT5350原理图及PCB
免责申明
【只为小站】的资源来自网友分享，仅供学习研究，请务必在下载后24小时内给予删除，不得用于其他任何用途，否则后果自负。基于互联网的特殊性，【只为小站】 无法对用户传输的作品、信息、内容的权属或合法性、合规性、真实性、科学性、完整权、有效性等进行实质审查；无论 【只为小站】 经营者是否已进行审查，用户均应自行承担因其传输的作品、信息、内容而可能或已经产生的侵权或权属纠纷等法律责任。
本站所有资源不代表本站的观点或立场，基于网友分享，根据中国法律《信息网络传播权保护条例》第二十二条之规定，若资源存在侵权或相关问题请联系本站客服人员，zhiweidada#qq.com，请把#换成@，本站将给予最大的支持与配合，做到及时反馈和处理。关于更多版权及免责申明参见 版权及免责申明
本站所有资源不代表本站的观点或立场，基于网友分享，根据中国法律《信息网络传播权保护条例》第二十二条之规定，若资源存在侵权或相关问题请联系本站客服人员，zhiweidada#qq.com，请把#换成@，本站将给予最大的支持与配合，做到及时反馈和处理。关于更多版权及免责申明参见 版权及免责申明