issues
search
senlinuc
/
caffe_ocr
主流ocr算法研究实验性的项目,目前实现了CNN+BLSTM+CTC架构
1.26k
stars
535
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
请问仅CPU如何编译呢?好像没有提供CommonSetting?
#94
ZipengFeng
closed
5 years ago
0
解决方案打开的问题
#93
gitmessage
opened
5 years ago
0
将fc1x层的节点数由5990变为36(因为我只识别数字和大写英文字母),无法训练,loss=nan
#92
HouBiaoLiu
opened
5 years ago
4
densenet-sum-blstm-full-res-blstm_train-val.prototxt只有看到一个标签,怎么样才能对多个标签进行训练
#91
shenjackson
opened
5 years ago
1
大神,你识别中文是Lexicon-based transcription还是Lexicon-free transcription
#90
bxhandhxb
opened
5 years ago
0
作者方便把训练的英文字典和数据也分享下吗?
#89
magneter
opened
5 years ago
1
Opencv版本
#88
ffyAI
opened
5 years ago
2
mean_value.txt
#87
panlanlan
opened
5 years ago
0
中文括号,导致链接错误了
#86
wangershi
opened
5 years ago
0
显示乱码
#85
xqpinitial
opened
5 years ago
0
数据库生成很好
#84
Caiyou1001
opened
5 years ago
4
移植到Linux
#83
FLHonker
closed
3 years ago
1
请问英文句子中空格怎样识别?
#82
Jeremyliu7
opened
5 years ago
3
The meaning of score
#81
nerddd
opened
5 years ago
0
请教大神 如何移植到Android上
#80
xiaomochen520
opened
5 years ago
2
denseBlock Layer中的initChannel 代表什么含义呢?
#79
ghost
closed
5 years ago
0
多值标签制作lmdb问题
#78
suntim
opened
5 years ago
3
ocr_test 有用到均值文件么,未什么我训练的时候准确率很高使用ocr_test测试的时候准确率很低
#77
ghost
closed
5 years ago
2
Check failed: axis_index < num_axes() (2 vs. 2) axis 2 out of range for 2-D Blob with shape 35 256 (8960)
#76
suntim
opened
5 years ago
3
验证机的准确率很高,测试集的准确率很低是怎么回事?
#75
ghost
closed
5 years ago
1
使用ImageData层和LMDB方式读取数据有区别吗?
#74
ghost
closed
5 years ago
1
使用orc_test测试的时候,输出label很多,且大部分是0是怎么回事
#73
ghost
closed
5 years ago
1
训练疑问,迭代两万次之后loss卡在30不降低了
#72
Levi-zhan
closed
5 years ago
7
DenseBlock 出错,cudnnTensorDescriptor_t’ does not name a type
#71
ghost
closed
5 years ago
1
warpctcloss层疑问
#70
prfans
closed
5 years ago
0
训练疑问
#69
sannyjo
opened
5 years ago
1
在中文数据集上crnn也是最高,能到99.26,你测试的数据集是30多万哪个么
#68
liny23
opened
6 years ago
18
是否ctc对连续数字的识别率不够高?
#67
ares89
closed
5 years ago
2
中英文准确率差别
#66
ecetang
opened
6 years ago
1
关于Resnet网络的训练微调的疑问
#65
bestzld
opened
6 years ago
0
compute_output_shape compute wrong output_shape value
#64
mbtolou
opened
6 years ago
1
CUDNN_CHECK(cudnnCreate(&handle_[g]));
#63
JaosonMa
closed
6 years ago
2
win7+gtx970 cuda8.0 编译报错 "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\bin\nvcc.exe" -gencode=arch=compute_20,code=\"sm_20,compute_20\" --
#62
JaosonMa
closed
6 years ago
1
目前版本是不是没有ctpn 只能识别一样文字
#61
xikunlun001
opened
6 years ago
2
请问下transpose层和permute层有什么区别吗
#60
shaoming20798
opened
6 years ago
1
warp_ctc_loss_layer.cpp:56] Check failed: N_ == label_seq*** Check failure stack trace: *** ->num() (128 vs. 64)
#59
prfans
closed
5 years ago
29
DenseBlock_layer.hpp 编译的时候出现错误
#58
Lesley96-11
opened
6 years ago
0
udnnDropoutDescriptor_t’ was not declared in this scope vector<cudnnDropoutDescriptor_t *> dropoutDescriptorVec;
#57
Lesley96-11
closed
6 years ago
0
DenseBlock_layer.cu(283): error: too many arguments in function call
#56
billyzju
opened
6 years ago
4
Test Net output 输出的有两个acc,分别指的是什么?
#55
AllenDuane
opened
6 years ago
1
编译时链接boost库版本问题。
#54
ForrestPi
opened
6 years ago
2
样本长度的问题
#53
weiweili123
opened
6 years ago
0
关于训练自己的样本,结果有问题
#52
xap1992
opened
6 years ago
3
请问一下,你用的LSTM层就默认就是有做残差的吗,残差是参考了那篇文章的?谢谢。
#51
Darwin84
opened
6 years ago
0
你好
#50
BowieHsu
opened
6 years ago
0
how to use only DenseBlock_layer
#49
zhimengfan1990
opened
6 years ago
0
语料库
#48
LiangHao92
opened
6 years ago
1
你好,在生成训练lmdb的时候出错,ReadImageToDatum的return true时报错
#47
hewumars
opened
6 years ago
0
CTC layer的实现是结合百度的warpCTC实现的吗?有没有可以直接使用的官方的ctcLayer
#46
AirFishWang
opened
6 years ago
0
楼主,为什么我识别后,总是只能识别第一个字,后面就识别不了了。
#45
GarryWan
closed
6 years ago
2
Previous
Next