Sergio Izquierdo
Sergio Izquierdo
Hi @grok-phantom Have you tried with GPU config options like `allow_growth` and `per_process_gpu_memory_fraction`. You can take a look at #93.
Closing due to inactivity.
Hi, So, the problem is about how to create a proper TF_TString tensor, right?
Hi @Amuqeet09 Sorry for the late late reply. I think this should not be an issue of cppflow, but rather the library of tensorflow. What libtensorflow version are you using?...
Hi @abucka , Sorry for the late reply. I have been looking for this in the C API and in tf.raw_ops and I didn't find how. Please, if you figured...
Glad that you found out how to achieve this!
Nice work :+1:
This is not required anymore, saved model is supported in the TF C API
Looks like you should not expand the dims in the second example, as it already has a batch dimension.
The library is intended to be used with c++17. Anyway, whoever is interested in c++14 can refer to this PR.