Fabian Boemer

@bcebere , thanks for the detailed report! Our current HEXL implementation/integration has focused on improving performance on AVX512-enabled machines. In particular, the recent Intel processors with the AVX512-IFMA52 instruction set...

We used to support encrypting model weights, but no longer do (https://github.com/NervanaSystems/he-transformer/commit/862e300f21d7b08b25ce5e878c1d18e6bb5c3414).

Does the `make install` step succeed? We expect `libhe_seal_backend.so` to appear in `$HE_TRANSFORMER/build/ext_ngraph_tf/src/ext_ngraph_tf/build_cmake/venv-tf-py3/lib` and `$HE_TRANSFORMER/build/external/lib`

So there are two orthogonal issues here: 1) The install; if the CryptoNets-Relu example is working, it seems the install worked okay. After the install, we expect *`$HE_TRANSFORMER/build/ext_ngraph_tf/src/ext_ngraph_tf/build_cmake/venv-tf-py3/lib` to contain...

As it turns out, `HE_SEAL is not supported` seems to be thrown whenever he-transformer throws an error, rather than printing the more descriptive reason. This is a reported issue with...

Thank you, @tiogoGit , @cjbumgardner for your reports. Here's more details on what the installation looks like at each step. Note, there is no `he_seal_backend.so`, just `libhe_seal_backend.so`, and `build/ext_ngraph_tf/src/ext_ngraph_tf/build_cmake/venv-tf-py3/lib` should...

Thanks @cjbumgardner for your detailed report again. A few notes: 1) Once merged, https://github.com/tensorflow/ngraph-bridge/pull/229 will hopefully return a more useful error message 2) I wouldn't recommend using older versions. There...

I believe TensorFlow simplifies simple expressions automatically. You could try more complex expressions rather than `(a+b)*a*b`. Maybe try ReLU or MaxPool, or square.

The choice of parameters is quite aggressive, resulting in low accuracy with some weights. You may try retraining the model. You may also modify `he_seal_ckks_config_N13_L7` to use more coefficient moduli,...

Hi Aleksandar, thanks for your interest. Unfortunately this library supports CPU only.