-
Epoch: [0][59460/102488] Time 0.087s (0.087s) Speed 366.4 samples/s Data 0.002s (0.009s) Loss 0.01915 (0.49964)
torch.Size([32, 512, 1, 41])
torch.Size([32, 512, 1, 41])
torch.Size([…
-
同学你好 ，我注意到测试代码中confidence_threshold=0.01，nms_threshold=0.45，这势必会造成高recall，低precision。是不是在localication的过程中，也使用了CRNN去纠正检测结果。我们组在尝试使用Tensorflow复现TextBoxes，目前我们尝试的参数confidence_threshold=0.3，nms_threshold=…
-
CRNN的输入是把图片的高度放缩到32，如果按照相同的比例将宽放缩，此时不同的图片的宽度是不一样的，这种情况应该如何训练模型？
-
**1. 使用环境（environment）**
- centos，cuda10.2，cudnn 8.2.0，tensorrt7.2.1
**2. Github版本**
- branch：master
- commit(optional):
**3. 详细描述bug 情况 (Describe the bug)**
TNN版本的crnn可以正常运行，尝试将其量化
…
-
File "/home/wzp/project/ppocr/lib/python3.6/site-packages/torch/nn/modules/module.py", line 722, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/wzp/project/torch/CRNN_svtr/…
-
Hi, thanks for your sharing!
I have joined the competition, and learned a lot from your repo.
But I have found that there are some files do not in the CRNN part , can you complete the supplem…
-
Hi @AlexeyAB ,
Thanks for ur contribution. Can you explain more about how to train tiny-occlusion track cfg? which pre-trained weight we can use and crnn ?
thanks in advance.
-
INVOICE-SCANNER]$ python app.py
Traceback (most recent call last):
File "app.py", line 5, in
from crnn_processor import get_html
File "/var/www/html/INVOICE-SCANNER/backend/ReceiptGenera…
-
hi,i want to run the demo.py with CPU ?what should i do ?
I change the ctpnport.py like this:
```
caffe.set_mode_cpu()
# caffe.set_device(cfg.TEST_GPU_ID)
```
and the ctpn works,but I change t…
-
我现在用的是DB+CRNN的流程做OCR 全是python版本的
python版本crnn trt的模型我试着测了一下耗时 发现还比ONNX慢 不清楚是不是C++和python的关系
python版本db trt我测试比onnx快了大概3倍 但是还是得200多ms（之前做过yolo检测 转trt大概20ms C++版本）
有点疑惑是python和C++会差那么多么（我也都是动态输入）
…