hello!
I want to make my own data using /scripts/convert_fc_to_tfrecords.py.
But when I used the data to train the flownetC, it showed "tensorflow.python.framework.errors_impl.InvalidArgumentError: Input to reshape is a tensor with 262144 values, but the requested shape has 131072''.
I checked the data size:input_a(5, 256, 256, 3),input_b(5, 256, 256, 3),flow(5, 256, 256, 2).
So I was wondering if some parameters need to be changed when the data size was changed?
hello! I want to make my own data using /scripts/convert_fc_to_tfrecords.py. But when I used the data to train the flownetC, it showed "tensorflow.python.framework.errors_impl.InvalidArgumentError: Input to reshape is a tensor with 262144 values, but the requested shape has 131072''. I checked the data size:input_a(5, 256, 256, 3),input_b(5, 256, 256, 3),flow(5, 256, 256, 2). So I was wondering if some parameters need to be changed when the data size was changed?