search

roatienza / deep-text-recognition-benchmark

PyTorch code of my ICDAR 2021 paper Vision Transformer for Fast and Efficient Scene Text Recognition (ViTSTR)
Apache License 2.0
284 stars 57 forks source link

Demo.py #18

Open Preethse opened 2 years ago

Preethse commented 2 years ago

I trained my model for japanese language. The validation accuracy goes upto 99.9% but when testing on test image it fails very badly. Can you share demo.py file so that I can check if I am doing everything correctly or not. Thank you

roatienza commented 2 years ago

Hi, thanks for using ViTSTR. If the validation accuracy is 99.9%, something is wrong. Unfortunately, I could not find the demo script I used for ViTSTR but it is simply a modified version of test.py. For ViTSTR to work, you need to make sure that the dataset is correct before training.

LeeBronOff23 commented 1 year ago

我为我的模型训练了日语。验证准确率高达 99.9%，但在测试图像上进行测试时，它非常失败。你能分享 demo.py 文件，以便我可以检查我是否正确地做所有事情。 谢谢 Hello, I also have the same problem. I have a high accuracy rate in other language training, but the test fails. Have you solved the problem now?