- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi to everyone, i have my neural network built in keras library that i converted to tensorflow model, and then i used the mvNCCompile to use it on the Movidius but i got an errore. Can you please explain me what is wrong? Thanks!
This is the code that i used to convert the keras model to tensorflow
model_file = "/home/ciao/Scrivania/model.json"
weights_file = "/home/ciao/Scrivania/weights.h5"
with open(model_file, "r") as file:
config = file.read()
K.set_learning_phase(0)
model = model_from_json(config)
model.load_weights(weights_file)
saver = tf.train.Saver()
sess = K.get_session()
saver.save(sess, "/home/ciao/Scrivania/TF_Model/tf_model")
fw = tf.summary.FileWriter('logs', sess.graph)
fw.close()
Then the error:
- Tags:
- Tensorflow
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
is it an SSD model?
you might be missing the config file
https://ncsforum.movidius.com/discussion/1588/ncsdk-2-10-option-tf-ssd-config
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
@robert_osmosis i worked on keras with a xception architecture, i dont have this config file how can i get it? (sorry but i'm learning on these things)
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
that architecture might not be supported with movidius
https://movidius.github.io/ncsdk/release_notes.html#network-notes
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
@robert_osmosis ok, thank you
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page