Closed codaibk closed 3 years ago
Thanks for the report. Could you add which code was being run, and copy-paste the stack trace included with the error message?
Thanks for your reply. I run differential privacy for emnist code. The error is shown in below pict:
@ZacharyGarrett Hi, is there any update?
It seems your original code has typo. It run smoothly after I changed the order of your code in function get_federated_datasets
of emnist_dataset.py
file.
I used this one:
feature_dtypes = collections.OrderedDict( label=tff.TensorType(tf.int32), pixels=tff.TensorType(tf.float32, shape=(28, 28)) )
instead of using your original code:
feature_dtypes = collections.OrderedDict( pixels=tff.TensorType(tf.float32, shape=(28, 28)), label=tff.TensorType(tf.int32) )
Might commit https://github.com/google-research/federated/commit/bea0484a47fe5ef3cce4c945b36ed93e7f34f06e have fixed this issue?
@codaibk As @ZacharyGarrett mentioned, commit bea0484a47fe5ef3cce4c945b36ed93e7f34f06e should fix the problem. Can you confirm or deny this?
@zcharles8 ya, I just only changed the emnist_dataset.py
and it fix the problem.
Hi, Thanks for your framework. I tried to run your code but get this error "TypeError: Values of type