ano_pred_cvpr2018
ano_pred_cvpr2018 copied to clipboard
Error on dataset download
I downloaded the ped2 data from the Baidu link which is in.rpm file format, then I converted the file into .tar.gz format and extracted it using the command tar -xvf ped1.tar.gz in the shell and the data is in .cpio format now.
Then I run the command to test on the save models. The code is running but I am getting output as:
test inputs = Tensor("dataset/strided_slice:0", shape=(1, 256, 256, 12), dtype=float32) test prediction gt = Tensor("dataset/strided_slice_1:0", shape=(1, 256, 256, 3), dtype=float32) testing = generator 2022-04-05 00:45:19.075695: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 AVX512F FMA Init global successfully! waiting for models...
optimal result and model = dataset = None, loss file = None, auc = -inf
I think the code is not able to get the data as the file format is .cpio. and after extraction of .tar.gz file it should be converted to image file format(.jpg or .tif) instead of .cpio format.
Hello, I'm facing the same mistake as you. Has your problem been solved?
Hello,
No, the problem is not solved.
Regards, Chinmaya Meher
On Mon, Mar 6, 2023 at 7:00 PM harveyzoka @.***> wrote:
Hello, I'm facing the same mistake as you. Has your problem been solved?
— Reply to this email directly, view it on GitHub https://github.com/StevenLiuWen/ano_pred_cvpr2018/issues/62#issuecomment-1456140660, or unsubscribe https://github.com/notifications/unsubscribe-auth/APHP5DCTUI5XIHLBZLUUQXLW2XRGNANCNFSM5SPT75ZQ . You are receiving this because you authored the thread.Message ID: @.***>