Open Kjos opened 6 years ago
Hello, I am training neural network using this code (cv-tricks.com-master\cv-tricks.com-master\Tensorflow-tutorials\tutorial-2-image-classifier\train) on my own data set of road traffic signs in which i have left and right road signs. I have changed the directory as follows:
classes = os.listdir("D:\My_tensorflow\data_set\Train") train_path="D:\My_tensorflow\data_set\Train" test_path="D:\My_tensorflow\data_set\Test"
when i run this code it gives this kind of out put and error: Number of files in Training-set: 0 Number of files in Validation-set: 0 error is: assert batch_size <= self._num_examples
Kindly help me to figure out this problem. Thanks
Hi, Thanks for putting this up on github. It got me started with tensorflow. However some comments; I think the dataset shows a lot of underfitting. I get a validation score of 0.5 at end, or even below. I connected tensorboard and the learning graph shows classic underfitting signs (straight learning rate). Perhaps it doesn't help that the images have wide ranging aspect ratios? I tried to insert horizontal flipped images as well, but it doesn't help that much.