zhiqwang/sightseq issues and pull requests

#22 - 能提供新的依赖版本么？

Issue - State: open - Opened by liuzhuang1024 about 4 years ago - 5 comments
Labels: enhancement

#21 - How is the picture processed in sequence_generate?

Issue - State: open - Opened by xiaoxiaokuye about 4 years ago - 1 comment
Labels: question

#20 - 关于加载预训练模型的问题

Issue - State: closed - Opened by xiiiiiiii over 5 years ago - 1 comment
Labels: question

#19 - Getting accuracy as 0.00

Issue - State: closed - Opened by Arvind-Kumar-Agrawal over 5 years ago - 1 comment
Labels: question

#18 - 有关loss变为nan的情况，我看了之前的解答，但还是想问问

Issue - State: closed - Opened by xiiiiiiii over 5 years ago - 2 comments
Labels: bug, question

#17 - 中文识别率不高是不是因为感受野的原因？

Issue - State: closed - Opened by xiiiiiiii over 5 years ago - 16 comments
Labels: bug, question

#16 - dimensions in forward pass

Issue - State: closed - Opened by soldierofhell over 5 years ago - 4 comments
Labels: question

#15 - Input size

Issue - State: closed - Opened by lvoegtlin over 5 years ago - 1 comment
Labels: duplicate, question

#14 - 中文识别率不高问题

Issue - State: closed - Opened by peiji1981 over 5 years ago - 20 comments
Labels: question

#13 - loss become inf , then Nan

Issue - State: closed - Opened by XiaogangYin over 5 years ago - 18 comments
Labels: bug, question

#12 - RuntimeError: CUDA error: an illegal memory access was encountered

Issue - State: closed - Opened by ronghui19 over 5 years ago - 7 comments
Labels: question

#11 - Questions about dataset object

Issue - State: closed - Opened by skurzhanskyi over 5 years ago - 4 comments
Labels: question

#10 - Help Needed

Issue - State: closed - Opened by SibtainRazaJamali over 5 years ago - 7 comments
Labels: question

#9 - 同一批测试数据，test-only 的accuracy和训练时的validate accuracy 差很多？

Issue - State: closed - Opened by peiji1981 over 5 years ago - 4 comments
Labels: bug

#8 - 训练结果在其他图片上的结果很差？

Issue - State: closed - Opened by bagpeng over 5 years ago - 3 comments
Labels: question

#7 - Must the training data be of equal length?

Issue - State: closed - Opened by carry-xz almost 6 years ago - 1 comment
Labels: question

#6 - Not found recurrent layer in model files

Issue - State: open - Opened by courao almost 6 years ago - 1 comment
Labels: enhancement

#5 - annotation file format for English data

Issue - State: closed - Opened by pp3993 almost 6 years ago - 2 comments
Labels: question

#4 - The vanilla cnn downsampling architecture cannot recover spatial information of a image

Issue - State: open - Opened by zhiqwang almost 6 years ago - 1 comment
Labels: enhancement

#3 - TypeError: 'DigitsBatchTrain' object is not iterable

Issue - State: closed - Opened by jianghonggang almost 6 years ago - 15 comments
Labels: bug

#2 - Error(s) in loading state_dict for CRNN:

Issue - State: closed - Opened by Bilgee almost 6 years ago - 1 comment
Labels: question

#1 - How to use it just for testing one normal image?

Issue - State: closed - Opened by oneTaken almost 6 years ago - 6 comments
Labels: good first issue

Ecosyste.ms: Issues

GitHub / zhiqwang/sightseq issues and pull requests