Hacker News Reader

> You need to create a quantized TensorFlow Lite model and then compile the model for compatibility with the Edge TPU. We will provide a cloud-based compiler tool that accepts your .tflite file and returns a version that's compatible with the Edge TPU.

I seriously hope that's not the only way they're expecting people to compile models for this particular TPU.

View on HN

As a person with access to their documentation I can confirm that this is currently the only way to compile model for this TPU. Also list of supported network architectures is very very short.

Well this is just begining I am sure they are going to expand its capability.

I'd rather they enable independent development of models on the hardware they're selling. This is about as useful as a high performance electric car you can only charge at authorised dealerships. Dealerships which have the unfortunate habit of closing down after a few years or so.

I bet there is some technical reason behind allowing just few architectures and I hope they will fix it in future releases. Currently you can't even run ResNet type of networks on it.

Well, one can hope.

Reply on news.ycombinator.com