ybdesire

Results 20 issues of ybdesire

https://stackoverflow.com/questions/47645280/how-to-do-predict-for-linearmodels

enhancement

目前该项目输出的是每一行文字的坐标,如果想得到每个字的坐标,定位到每一个汉字,该怎么做呢? 烦请分享下思路即可,谢谢!

目前该项目输出的是每一行文字的坐标,如果想得到每个字的坐标,定位到每一个汉字,该怎么做呢? 烦请分享下思路即可,谢谢!

自己构建的数据集,无法满足每个样本固定10个字符。 请问用什么方法/项目,能做到把您提供的百度网盘中的`Chinese_dataset\images`剪裁为固定10个字符的训练集呢?

What is the minimum V100 or A800 numbers we need to train this model if not considering the training time ?

请教一下:和不量化的7B/14B模型对比,72B模型量化后,效果怎么样呢?是量化的72B会差很多,还是完全没有可比性呢?

tesseract可以输出识别到的每个字的坐标,请问deep_ocr该怎么用才能输出字坐标呢? 谢谢

如何构建dataset.pkl这个数据?

### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior 比如,微调模型的配置如下 ``` --max_steps 3000 \ --save_steps 5\ ``` 这样保存的checkpoint就会从5, 10, 15,...

It's easier to just upload data to online tensorboard here http://projector.tensorflow.org/ to do data visualization. Is it possible to just upload data to local tensorboard as online tensorboard?

type:feature
plugin:projector