Issues
- 4
如何基于 GPT2增量训练或微调实现生成式任务
#401 opened by runningabcd - 0
- 0
couplet.txt怎么生成的,格式是什么样的
#399 opened by Liufeiran123 - 0
预训练的分类和微调中的分类微调有什么区别
#398 opened by lizhipengpeng - 0
model zoo里的一些预训练模型找不到了
#391 opened by StudentxWan - 0
支持deepspeed的分支
#390 opened by zhang2010hao - 2
请教下有gpt2的微调的相关文档么?
#353 opened by ucas010 - 3
- 0
- 0
mBART?
#363 opened by 5i-wanna-be-the-666 - 0
请问wwm不适用于生成mlm+nsp格式的数据吗?
#362 opened by dr-GitHub-account - 0
预训练问题
#361 opened by Clearloveplus7 - 0
- 0
seq2seq模型验证时如何使用BLUE这些指标,我看里面都是混淆矩阵
#358 opened by panbo-bridge - 1
作者你好,请问small_config.json这个配置文件在哪呢
#357 opened by leiqing110 - 0
T5模型预训练问题
#356 opened by zhangzai666 - 0
问一下t5预训练模型如何推理
#355 opened by zhangzai666 - 0
请问有没有UER的t5v1.1和google的mT5中中文能力的对比呀?
#354 opened by nameless0704 - 0
如何训练seq2seq 的unilm
#352 opened by zhihao-chen - 0
Is it possible to get embedding directly
#351 opened by ChongruiYang - 1
- 1
- 0
中文特定领域预训练数据集规模
#348 opened by dr-GitHub-account - 1
- 1
使用第三方预训练模型的参数设置问题
#340 opened by kxy-cheng - 1
UER支持直接加载mT5吗?
#343 opened by blueseasky - 1
LSTM的预训练模型的分词用的是什么
#346 opened by wanyuks - 0
请问user-py的 tokenizer是如何保存和转换的
#345 opened by ray075hl - 0
cosine similarity为什么会出现负值?用sentence_transformers的util.cos_sim()调用uer/sbert-base-chinese-nli
#344 opened by peter65374 - 6
convert_t5_from_uer_to_huggingface.py运行报错
#339 opened by ColaFei - 6
How should I preprocess training data samples when I use HuggingFace Transformers T5?
#338 opened by ShaneTian - 1
gate_cnn 模型GPU多卡训练报错
#337 opened by TestNLP - 3
相同条件,每次输出的结果均不同
#336 opened by bobo0810 - 3
在训练BERT时, Loss突然增大且模型无法继续学习
#335 opened by xlxwalex - 3
微调 finetune/run_classifier_siamese.py 模型不收敛。
#330 opened by yuanfengning - 1
- 1
不支持M1芯片
#332 opened by Eggwardhan - 0
finetune/run_classifier.py 关于文本对训练中seg的赋值
#334 opened by husheng-liu - 0
xlm-roberta-base转换为uer格式后能直接用于下游任务吗?
#333 opened by jeave - 1
convert_bert_from_uer_to_huggingface script does not generate config.json file
#329 opened by chenweiyj - 2
预训练时加载huggingface T5模型报错
#328 opened by fade-color - 1
ELECTRA预训练模型转换至UER格式
#325 opened by sl403 - 2
pegasus 训练的dataset是不是和原文章不相同啊?
#331 opened by CheaSim - 3
- 0
请问如果要训练英文版的T5,词表文件去哪里找呢
#326 opened by fade-color - 6
新代码多机多卡会卡死
#324 opened by seeledu - 1
deepspeed单机/多机显存占用一样问题
#311 opened by cxfzzj - 2
请问run_classifier_multi_label.py的label数据格式是什么?
#320 opened by JYS-99 - 3
请问一般BERT预训练的acc_mlm大概训练完后能到多少
#313 opened by xlxwalex - 0
Is the [SEP] token missing here at the end?
#314 opened by raytien