Federico Jensen Carrasco
Federico Jensen Carrasco
It would be great if you can incorporate this :)
Same error
@AutumnSun1996 I'm working on it in my spare time. I am having doubts if the transition process I mentioned in the guide from PyTorch to ONNX generates lower accuracy and/or...
@Itaybre python 3.9.9 torch 1.10.1 torchvision 0.11.2 onnx 1.11.0 onnxruntime 1.11.1 These are the versions of the main packages I have installed
Hi everyone, thanks for the feedback. I have been incredibly busy the last 3 weeks. I finally have some free time. I will check again the bug to find the...
Okay guys I found the source of the error mentioned by @Itaybre. The problem is that "torch.onnx.export", in case of the recognition model, only works when EasyOCR is running in...
I have created a [new issue](https://github.com/JaidedAI/EasyOCR/issues/786) where I have made available the ONNX version of the EasyOCR models for all languages. Feel free to download and use them.
I have published a branch in [this fork](https://github.com/Kromtar/EasyOCR-ONNX/tree/easyocr_onnx) where you can find the whole process using containers. You can see the readme to understand how to use it.
@long-senpai I don't really know why this happens in the conversion process. I think that ONNX, when optimizing the model, discovers that the weights provided by input 2 are unnecessary;...
In the [examples folder](https://github.com/govorox/SSLClient/tree/master/examples/Esp32) you can find fully functional codes.