@Tome_at_Intel I had to upgrade to TF 1.5 to get rid of the training features.
But now I see:
/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py:858: DeprecationWarning: builtin type EagerTensor has no __module__ attribute
@Tome_at_Intel: I understand the exception what I don't understand is what exactly am I dong wrong in converting my Keras Model to Tensorflow. I don't set FusedBatchNorm explicitly. Also, you didn't address the patch I posted above to the TensorFlow…
Note, here is some code I've used to translate Keras to TF (taken from a StackOverflow post I believe):
import tensorflow as tf
from tensorflow.python.framework.graph_util import convert_variables_to_constants
from keras import backend as…
Those three links are two PBs and a meta. All of them are translated Keras -> TF models.
Note to get the 'FusedBatchNorm' issue I had to patch the NCSDK like so:
-- TensorFlowParser.py 2018-01-29 11:37:22.912714875 +0000
I also see the following error on some other models:
[Error 5] Toolkit Error: Stage Details Not Supported: FusedBatchNorm inputs mean and variance are not defined. The graph is not created for inference.
This was of course me converting a Keras m…
I really wish there was more of a timeline on a fix for this. Can we at least get a patch release of 1.07.07 to address this specific issue?
My dongle is useless to me right now. I have tried everything to get it to work to no avail.
Thanks Tome. Do you have an ETA for a fix? It is sorta a show stopper for many of us.
If you need any debug information or would like to try something on a system that is effected by this bug, please don't hesitate to reach out.
Technically speaking, the front ports on my workstation are connected to an internal USB 2.0 hub on the motherboard. Why isn't that good enough?
I'd really, really like to get this to work with my existing hardware. I don't think that is too much t…
So it looks like other folks have reported this issue here as well:
It seems like the dongle is suppose to be enumerated in 2.0 mode until a graph is uploaded, and then it switch…