Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / meijieru/crnn.pytorch issues and pull requests

#92 - why my accuracy is always 0?

Issue - State: open - Opened by MingChaoXu almost 7 years ago - 49 comments

#91 - how to make dataset

Issue - State: closed - Opened by MingChaoXu almost 7 years ago - 4 comments

#90 - how to use lexicon in this project?

Issue - State: closed - Opened by fnzhan almost 7 years ago - 1 comment

#89 - AttributeError: 'module' object has no attribute 'gpu_ctc'

Issue - State: closed - Opened by cbzeng110 almost 7 years ago - 1 comment

#88 - sampler is mutually exclusive with shuffle

Issue - State: closed - Opened by axzhcode almost 7 years ago - 1 comment

#87 - KeyError: 'unexpected key "module.cnn.conv0.weight" in state_dict'

Issue - State: closed - Opened by LW-CVer almost 7 years ago - 2 comments
Labels: duplicate

#86 - Can I use any img_w in this crnn?

Issue - State: closed - Opened by zoe218 almost 7 years ago - 2 comments

#85 - torch to pytorch: KeyError: 'b'

Issue - State: closed - Opened by SylviaHolic almost 7 years ago - 1 comment

#84 - AssertionError: the height of conv must be 1

Issue - State: closed - Opened by gachiemchiep almost 7 years ago - 8 comments

#83 - how can I bind warp-ctc to pytorch?

Issue - State: closed - Opened by LiangHao92 almost 7 years ago - 1 comment

#82 - Test Run on images with low resolution and low Quality

Issue - State: closed - Opened by arsalan993 almost 7 years ago - 1 comment

#81 - Python and Pytorch version

Issue - State: closed - Opened by arsalan993 almost 7 years ago - 3 comments

#80 - Getting this error on Anaconda Python

Issue - State: closed - Opened by arsalan993 almost 7 years ago - 2 comments

#79 - In training phase,I need to resize img.width to 100 or 32*(img.width/img.height)?

Issue - State: closed - Opened by jycloud almost 7 years ago - 1 comment

#78 - Simple image path argument in demo script

Pull Request - State: open - Opened by MaxwellRebo almost 7 years ago

#77 - performance measure ???

Issue - State: closed - Opened by peternara almost 7 years ago

#76 - have not found convert_dataset.py

Issue - State: closed - Opened by dq-soulie almost 7 years ago - 1 comment

#75 - How to use the newly trained model?

Issue - State: closed - Opened by PsyDog5hao almost 7 years ago - 2 comments

#74 - Why 26?

Issue - State: closed - Opened by rohun-tripathi almost 7 years ago - 1 comment

#73 - LSTM

Issue - State: closed - Opened by lngao about 7 years ago - 1 comment

#72 - Is this software using a lexicon ?

Issue - State: closed - Opened by emalti about 7 years ago - 1 comment

#71 - In the code, there are different parts of the paper.

Issue - State: closed - Opened by peternara about 7 years ago - 2 comments

#70 - Finetuning on my dataset

Issue - State: closed - Opened by wahrheit-git about 7 years ago - 3 comments

#69 - Ask some questions about accuray

Issue - State: closed - Opened by wineternity about 7 years ago - 7 comments

#68 - Training dataset for the Dropbox model

Issue - State: closed - Opened by abuvaneswari about 7 years ago - 3 comments

#67 - Imbalanced classes

Issue - State: closed - Opened by skalinin about 7 years ago - 4 comments

#66 - KeyError: 'rnn.1.embedding.weight'

Issue - State: open - Opened by Amore-HDU about 7 years ago - 9 comments

#65 - How to training with variable length?

Issue - State: closed - Opened by zhangcliff about 7 years ago - 1 comment

#64 - AttributeError: 'module' object has no attribute 'gpu_ctc

Issue - State: closed - Opened by Amore-HDU about 7 years ago - 2 comments

#63 - Multi-gpu in training crnn model.

Issue - State: closed - Opened by SnailTyan about 7 years ago - 7 comments
Labels: bug

#62 - AssertionError: texts with length: 1664 does not match declared length: 6656

Issue - State: closed - Opened by haikuoyao about 7 years ago - 9 comments

#61 - Updated to work with pytorch 0.2.

Pull Request - State: closed - Opened by Weihnachtswurst about 7 years ago - 1 comment

#60 - unexpected key "module.cnn.conv0.weight" in state_dict

Issue - State: closed - Opened by Bilgee about 7 years ago - 3 comments

#59 - key error when running demo with fine-tuned model?

Issue - State: closed - Opened by lulurelax about 7 years ago - 8 comments

#58 - validation error happens during training phase

Issue - State: closed - Opened by white2018 about 7 years ago - 13 comments

#57 - Warp-CTC vs CTC ?

Issue - State: closed - Opened by ibrahimsoliman97 about 7 years ago - 2 comments

#56 - How CTC deals with successive doubled characters ?

Issue - State: closed - Opened by ahmedmazari-dhatim about 7 years ago - 5 comments

#55 - ValueError(‘sampler is mutually exclusive with shuffle’)

Issue - State: closed - Opened by ahmedmazari-dhatim about 7 years ago - 3 comments

#54 - Any ideas on variable length input images rather than resizing it ?

Issue - State: closed - Opened by rremani over 7 years ago - 4 comments

#53 - Use case of detection of multi words and sentences in a image

Issue - State: closed - Opened by mit456 over 7 years ago - 4 comments

#52 - text = torch.IntTensor(opt.batchSize * 5)????

Issue - State: closed - Opened by lxj0276 over 7 years ago - 1 comment

#51 - the net doesn't work,when I run the net, it stopped by the flowing informations

Issue - State: closed - Opened by ghost over 7 years ago - 7 comments

#50 - training on custom dataset

Issue - State: open - Opened by azamshoaib over 7 years ago - 13 comments

#49 - add gen_image, img2lmdb

Pull Request - State: closed - Opened by ZhangXinNan over 7 years ago - 5 comments

#48 - can not find the training data directory

Issue - State: closed - Opened by lxj0276 over 7 years ago

#47 - training data format?

Issue - State: closed - Opened by lxj0276 over 7 years ago - 1 comment

#46 - error: invalid argument '-std=c++11' not allowed with 'C/ObjC'

Issue - State: closed - Opened by ZhangXinNan over 7 years ago - 1 comment

#45 - add .gitignore, add "if torch.cuda.is_available():" in demo.py

Pull Request - State: closed - Opened by ZhangXinNan over 7 years ago - 2 comments

#44 - crnn

Issue - State: closed - Opened by ahmedmazari-dhatim over 7 years ago

#43 - GPU

Issue - State: closed - Opened by ahmedmazari-dhatim over 7 years ago

#42 - Confusion matrix for CRNN

Issue - State: closed - Opened by ahmedmazari-dhatim over 7 years ago - 8 comments

#41 - when using Multi-GPU training , get "ValueError: result of slicing is an empty tensor"

Issue - State: open - Opened by XuLiangFRDC over 7 years ago - 6 comments
Labels: bug

#40 - Why preds variable contains negative values?

Issue - State: closed - Opened by hellbago over 7 years ago - 14 comments

#39 - RuntimeError when launching demo.py

Issue - State: closed - Opened by hellbago over 7 years ago - 1 comment

#38 - Issue with LMDB file while learning

Issue - State: closed - Opened by kergann over 7 years ago - 10 comments

#36 - using gpu error:"'module' object has no attribute 'gpu_ctc'"

Issue - State: closed - Opened by lg12170226 over 7 years ago - 4 comments

#35 - how to run with cpu

Issue - State: closed - Opened by exgile over 7 years ago - 7 comments

#34 - when using Multi-GPU training , get "all tensors must be on devices[0]"

Issue - State: closed - Opened by Lzc6996 over 7 years ago - 10 comments

#33 - why the size of images in test are (100,32) instead of (128,32)

Issue - State: closed - Opened by Lzc6996 over 7 years ago - 1 comment

#32 - Error while converting crnn_demo_model.t7 into a .pth model

Issue - State: closed - Opened by rayush7 over 7 years ago - 11 comments

#30 - how can l fine tune the model ?

Issue - State: closed - Opened by ahmedmazari-dhatim over 7 years ago

#29 - Loss going to infinity after one minibatch

Issue - State: closed - Opened by rohitgajawada over 7 years ago - 12 comments

#28 - cublas runtime error： the GPU program failed to execute

Issue - State: closed - Opened by tlatlbtle over 7 years ago - 5 comments

#27 - How can l get the probability of the sequence outputted by CRNN ?

Issue - State: closed - Opened by ahmedmazari-dhatim over 7 years ago - 6 comments

#26 - Training got stuck with cuda

Issue - State: closed - Opened by codyhan over 7 years ago - 1 comment

#25 - raise StopIteration error in crrn_main.py DUE to max_iter

Issue - State: closed - Opened by ahmedmazari-dhatim over 7 years ago - 6 comments

#24 - Why only use lowercase letters?

Issue - State: closed - Opened by xinghedyc over 7 years ago - 2 comments

#21 - Training loss and val loss decrease normally, but accuracy is always 0

Issue - State: closed - Opened by yisongbetter over 7 years ago - 7 comments

#20 - torch.FloatTensor(opt.batchSize, 3, opt.imgH, opt.imgH) in crnn_main.py

Issue - State: closed - Opened by ahmedmazari-dhatim over 7 years ago - 6 comments

#19 - Issue while training

Issue - State: closed - Opened by abhiML over 7 years ago - 4 comments

#18 - AssertionError: the height of conv must be 1

Issue - State: closed - Opened by abhiML over 7 years ago - 12 comments

#14 - AttributeError: module 'utils' has no attribute 'data_parallel'

Issue - State: closed - Opened by rohitgajawada over 7 years ago - 4 comments

#11 - Does the model requires png images to perform well ?

Issue - State: closed - Opened by ahmedmazari-dhatim over 7 years ago

#10 - Issues related to dataset.resizeNormalize((100, 32))

Issue - State: closed - Opened by ahmedmazari over 7 years ago - 6 comments

#7 - can you give train image and label for example?

Issue - State: closed - Opened by bear63 over 7 years ago - 1 comment

#6 - License?

Issue - State: closed - Opened by amitdo over 7 years ago - 2 comments

#5 - Did you achieve the same accuracy as the original crnn?

Issue - State: closed - Opened by melody-rain over 7 years ago - 8 comments

#4 - cnn height equals 1 assertion error

Issue - State: closed - Opened by daobilige-su over 7 years ago - 3 comments

#3 - sixlmdp_train_data_ordered module is missing

Issue - State: closed - Opened by daobilige-su over 7 years ago - 3 comments

#2 - About the process of training a new model

Issue - State: closed - Opened by daobilige-su over 7 years ago - 10 comments

#1 - nn.parallel.data_parallel(self.cnn, input, gpu_ids) giving error

Issue - State: closed - Opened by anindyasarkarIITH over 7 years ago - 11 comments