bojone / keras_recompute
saving memory by recomputing for keras
☆37Updated 4 years ago
Alternatives and similar repositories for keras_recompute:
Users that are interested in keras_recompute are comparing it to the libraries listed below
- bert-of-theseus via bert4keras☆31Updated 4 years ago
- 中文 预训练 ELECTRA 模型: 基于对抗学习 pretrain Chinese Model☆139Updated 4 years ago
- XLNet: Generalized Autoregressive Pretraining for Language Understanding 论文的中文翻译 Paper Chinese Translation!☆49Updated 5 years ago
- tensorflow version of bert-of-theseus☆62Updated 4 years ago
- 2019达观杯实体识别☆19Updated 5 years ago
- 2019中国高校计算机大赛——大数据挑战赛 第一名解决方案☆41Updated 4 years ago
- 高性能小模型测评 Shared Tasks in NLPCC 2020. Task 1 - Light Pre-Training Chinese Language Model for NLP Task☆57Updated 4 years ago
- pytorch版bert权重转tf☆21Updated 4 years ago
- 第三届魔镜杯大赛, 智能客服聊天机器人真实数据, 提高智能客服的识别能力和服务质量☆38Updated 6 years ago
- Memory for Knowledge Graph, using Neo4j. 知识图谱存储与查询。☆45Updated 2 months ago
- DistilBERT for Chinese 海量中文预训练蒸馏bert模型☆91Updated 5 years ago
- ☆22Updated 6 years ago
- adafactor optimizer for keras☆20Updated 3 years ago
- ☆44Updated 5 years ago
- UNF(Universal NLP Framework)☆70Updated 4 years ago
- Adversarial Training for NLP in Keras☆46Updated 4 years ago
- lightweighted deep learning inference service framework☆39Updated 3 years ago
- Keras implement of Lazy optimizer☆21Updated 5 years ago
- CLUE baseline pytorch CLUE的pytorch版本基线☆74Updated 4 years ago
- 2019达观杯信息提取第5名代码☆20Updated 5 years ago
- This is a collection of convenient methods for data science competition.☆42Updated 4 years ago
- CCL2018客服领域用户意图分类冠军1st方案☆147Updated 2 years ago
- this is roberta wwm base distilled model which was distilled from roberta wwm by roberta wwm large☆65Updated 4 years ago
- 首届中文NL2SQL挑战赛复赛方案,评估数据集acc:0.85 复赛线上成绩: 0.833 Top15☆67Updated 4 years ago
- “达观杯”文本智能处理挑战赛☆54Updated 6 years ago
- 24*2个预训练的小型BERT模型,NLPer炼丹利器☆52Updated 4 years ago
- Dilation Gate CNN For Machine Reading Comprehension☆17Updated last year
- Adversarial Attack文本匹配比赛☆42Updated 5 years ago
- This is our solution for WSDM - DiggSci 2020. We implemented a simple yet robust search pipeline which ranked 2nd in the validation set a…☆62Updated 4 years ago
- 高校赛2019 文本点击预测☆42Updated 5 years ago