Closed neutr0nStar closed 2 years ago
all steps are wrote in README.md
Thank you for replying, but the tensorflow resnet pretrained weights links do not work, please fix them. And for trained weights, it says "Please download trained models by this project", where can I find these, are there any examples available on the internet ?
please use pretrain weight from MxNet
trained models
By the way, I have not enough space at google drive, so you can tell me which trained model that you want to download if you can not sign in baidu drive, I will send it to your email.
I can totally understand, after all drive space is limited, and often quite less. When I am trying to open the Baidu Drive link, it's asking for some "Extraction code", am I doing something wrong ?
I tried downloading, but I am getting very less speed. I will mail you, please reply on that mail with trained model.
I am trying to run this repo on the model you shared with me, I have pasted the model in _RotationDetection/output/trainedweights. I am using google colab. When I am trying to create tfrecord, it gives an error saying /data/dataset/DOTA/val/images/images. And when I run for training and testing, I get the error no CUDA-compatible device found even though I am using GPU as a hardware accelerator.
Ok, so despite giving these errors, it generated output files. And that has helped me complete my project. As a concluding note, I would like to thank you for making this amazing repo and being so helpful.
Sorry if my question is irrelevant, but this is what I am instructed to do as an assignment. I am instructed to use this repo for detecting the orientation of various building in an image. Please guide me for the same.