Issues
- 0
希望能出一份中文档
#109 opened by leoFitz1024 - 0
how to fine tuning with pre_trained model
#119 opened by WY19940327 - 0
tensorflow版本target端为什么只在结束加eos,却没有在开始加bos。
#118 opened by LJLQ - 0
- 0
- 0
- 0
- 1
你好,请问训练出现KeyError: '<unk>'是怎么回事
#88 opened by edwardelric1202 - 1
TypeError: Can't instantiate abstract class MapDataset with abstract methods _inputs, set_inputs
#104 opened by zhuchenxi - 1
报错:TypeError: Expected 'Iterator' as the return annotation for `__iter__` of Dataset, but found thumt.data.iterator.Iterator
#113 opened by leileilin - 2
训练时没有生成eval文件夹,也没有在日志中输出验证信息
#112 opened by edwardelric1202 - 2
- 2
pytorch version ? Providing a bool or integral fill value without setting the optional `dtype` or `out` arguments is currently unsupported. In PyTorch 1.7,
#103 opened by anbo724 - 2
- 1
Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!
#106 opened by treeson-li - 4
use cpu to inference
#95 opened by qpzhao - 0
Question about translating with CPU
#105 opened by T2shen - 0
模型训练无法收敛
#102 opened by baoyu-yuan - 0
translator.py生成了空的文档,程序无报错
#101 opened by Linxia-MUC - 1
- 0
get_relevance出现cast float to string报错
#98 opened by fringe-k - 5
about the time for train a model
#97 opened by Rooders - 3
- 1
batch_size 10G GPU 单机最大能做到多少
#89 opened by caoyuji1986 - 1
预训练模型
#90 opened by duguiming111 - 1
- 0
- 0
wmt14 en-de
#92 opened by hljjjmssyh - 1
如果中断后继续我上个检查点训练
#91 opened by onoff888 - 8
multiple GPUs training with pytorch
#86 opened by jennifer1995 - 1
Can I output the translation every 1000 step?
#87 opened by yinghy18 - 2
如何正确的加入预训练的词向量
#84 opened by orangefly0214 - 0
distributed training
#85 opened by shawnkx - 1
- 2
- 1
Bugs in bin/scorer.py
#81 opened by zhanghuimeng - 1
- 1
I do not find position_info_type in hyper parameter list in pytorch version thumt
#79 opened by shawnkx - 2
MRT tends to deteriorate the performance while fine tuning a pre-trained Transformer.
#78 opened by yongchanghao - 1
What's the suggested loss_scale value?
#72 opened by Felixgithub2017 - 0
关于update_cycle
#77 opened by ElliottYan - 1
checkpoint averaging error.
#74 opened by Felixgithub2017 - 1
UnicodeDecodeError: 'charmap' codec can't decode byte 0x81 in position 14:
#76 opened by HassanNaeemjutt - 8
Has sb. trained the transformer model on WMT14 en-de and test on newstest2014?
#71 opened by minorfox - 18
en2zh的实验,在decode阶段,出现空行
#70 opened by wwy510553871 - 6
A question of the replacement of "@@"
#69 opened by minorfox - 1
请问可以提供一下数据吗,谢谢
#68 opened by jichangzhen - 1
IndexError: tuple index out of range
#67 opened by zhangzhen999 - 1
请问MRT有其他框架的实现吗?
#66 opened by zhajiahe - 9
关于bleu 在wmt2017 de-en
#65 opened by zl1300012994