predict the test pic
Hi, neka-nat, Have you wrote code for testing ? If so , can you upload it ?
I have finished the training for master branch, I try to use "preds = com_model.predict(test_img)" to test my pictures after each training epoch, test_img.shape=(1,256,256,3). But the test results are too bad which is totally different from the result of training. Give me some help please. How to predict images with arbitrary mask shape.
I have trained the model with epoch number 100 and 36500 data, but in the paper they have trained with 8 million data and half a million epochs. They said it took two month. Therefore, one solution is simply to increase the number of epochs 100->500000.
I will also implement test code and check test results.
Oh ,thanks,wating for your code and result .
Hi, neka-nat, Have you complete the testing part? Badly need your test code
I added code for estimation. Please try to put data in the test folder and execute the script.