meijieru / crnn.pytorch

Convolutional recurrent network in pytorch

meijieru/crnn.pytorch Issues

during multi-language model training, accuracy of English/numbers is worse than Chinese characters, how to improve?
Updated 3 months ago
TypeError: can't pickle Environment objects?
Updated a year ago1
Why My CTC Loss is always Nan or Inf value?
Updated 2 years ago6
What is the difference between warp_ctc_pytorch.ctcloss and torch.nn.ctcloss?
Updated 2 years ago3
【劝退】大家不要在这里浪费时间了，这套代码bug 太多，也没有人维护。
Closed 4 years ago19
出现报错RuntimeError: CUDA error: an illegal memory access was encountered
Updated 2 years ago3
RuntimeError: CUDA error: an illegal memory access was encountered
Updated 2 years ago1
训练数据集用的是什么？？？
Updated 2 years ago1
Something wrong in class resizeNormalize(object)
Closed 3 years ago
thi project support python3?
Updated 3 years ago
关于模型结构最后两层的BN层
Updated 3 years ago1
Why is the pooling layer in the network different from the 1 * 2 pooling in the original paper?
Updated 3 years ago
The loss occurred in Nan's case,What is the reason?
Updated 3 years ago1
训练自己的数据时报错：an illegal memory access was encountered
Updated 3 years ago1
CRNN过拟合求解The Problem Of Overfitting
Closed 3 years ago
The Problem Of Over fitting过拟合问题
Updated 3 years ago
raise UnidentifiedImageError( PIL.UnidentifiedImageError: cannot identify image file <_io.BytesIO object at 0x0000024631A5E360>
Updated 3 years ago1
Is it normal for the model to predict the same string for the whole train batch and val?
Closed 3 years ago
Utility of method oneHot in utils.py
Updated 3 years ago
训练icdar2015
Updated 3 years ago
训练ICDAR2003, ICDAR2013
Updated 3 years ago4
loss and accuracy are both 0
Updated 3 years ago6
errors about training your own model
Updated 3 years ago2
不收敛，loss下降到15左右就震荡不动了
Updated 4 years ago3
Using this model only for digits classification
Updated 4 years ago
ImportError: /usr/lib/libtorch_cpu.so: undefined symbol: _ZNK3c1010TensorImpl35compute_channels_last_contiguous_2dEv
Updated 4 years ago1
效果不好
Updated 4 years ago3
Height of image to 1
Updated 4 years ago
loss值是特别大，但是还不知道那么下降，怎么解决这个问题？
Updated 4 years ago
About loss
Updated 4 years ago1
eval mode has bad performance 测试与训练模式性能差别很大
Updated 4 years ago3
TypeError: function takes exactly 5 arguments (1 given)
Updated 4 years ago2
assert index <= len(self), 'index range error' AssertionError: index range error
Updated 4 years ago4
对train.py的几点疑问
Updated 4 years ago
train.py文件的第98行代码的疑惑
Updated 4 years ago1
支持不同样本训练吗
Updated 4 years ago
TypeError: a bytes-like object is required, not 'NoneType'
Closed 4 years ago2
How do I count the parameters of the model to calculate the complexity of the model?
Updated 4 years ago1
如果需要训练图片的尺度为64会报错
Updated 4 years ago1
HELP！如果我想批处理测试数据集，我该怎么修改啊
Updated 4 years ago1
How to view the score of character recognition, that is, the probability value of recognition result
Updated 4 years ago
Is there anything wrong about the shape of CRNN modual
Updated 4 years ago
图像的宽度
Closed 5 years ago
训练的trainroot具体是什么啊
Updated 5 years ago1
Speedup inference in strLabelConverter
Updated 5 years ago
Meaning of loadData(v, data) function in utils.py file.
Updated 5 years ago1
中文字符无法识别
Updated 5 years ago1
why self.alphabet = alphabet + '-' ?
Updated 5 years ago3
Input to train.py
Closed 5 years ago
定常训练，不定长测试
Closed 5 years ago1