Upcoming maintenance
Dear Customers and Partners.
This website will be undergoing scheduled maintenance on June 14, 2023. Please be aware there may be disruption to the developer portal website and associated services during the scheduled maintenance period.
This upgrade is essential to ensure the continued performance, reliability, and security of Developer World.
We apologize for any inconvenience.
Difficulties in tensorflow Lite model deployment
-
We tried loading a simple custom trained RNN model with 80 neutrons in spresense sdk and the following error popped up, how can we deploy the model succesfully?
-
@satish same issue someone kindly help
-
It looks like the current version of TensorFlow LM from Spresense SDK does not support your models.
TensorFlow LM will be updated in the next SDK release. I think this will solve this problem. You can try to build the latest TensorFlow LM (https://github.com/tensorflow/tflite-micro) on your computer and see if your model works. If it works, then the next SDK release will solve your problem.
Best Regards,
Kamil Tomaszewski