leimao/Frozen-Graph-TensorFlow

read values

AshishGusain17 opened this issue · 5 comments

Hey, Can you tell how to read the values like weights and gradients using this frozen graph.
We also have other model.ckpt.data .meta .index files.
Is there any way?

You could figure out a way to load checkpoint models as a Keras model, then you could try to export a frozen model using the methods I described.

Is there even a way to load inference models as keras models. If so, can you share some articles or docs file link

It is not possible to load the frozen model to Keras as far as I know.

So, how can I load weights to keras model.
I have tensorflow model available i.e. model.ckpt,data,frozen graph,checkpoint.
If not load, how can I get the values of weights and gradients in each layer

HI @leimao, thanks for the documentation, I wanted to ask you if there is any way to make batch predictions using the freeze model, similar to how the predict() function is used with the 'batch_size=' parameter