whatdhack

@dariopavllo thanks. So as I understand there are 3 steps in training from scratch - convmesh , inverse rendering and GAN training in order. So for my custom image sets,...

@dariopavllo , thanks. The first step, convmesh, is a stripped down version of cmr as you pointed out in the paper, hence that alone can theoretically generate the geometry and...

@ft3020997 , good find indeed. However in my case I have to use other MPIs - Intel MPI, oneccl, MPICH etc. These are all pre-installed in the system. Cmake I...

@tgaddair , thanks, Is there environment variable knobs to set these compilers ?

This I think should be very high priority ( at the least FP16) , otherwise the case for TFS becomes weak.

How do I turn on AMP in serving ? I have observed 50% improvement in processiong time with fp16 over fp32 without any noticeable change in accuracy. Reduced precision is...

Is there a way to do the following in TFS ? ``` config = tf.ConfigProto() config.graph_options.rewrite_options.auto_mixed_precision = 1 sess = tf.Session(config=config) ```

I just ran some tests on a MaskRCNN Saved Model in nvcr.io/nvidia/tensorflow:20.03-tf1-py3. TF_ENABLE_AUTO_MIXED_PRECISION seems to work very well for inference - requires less memory and speeds up significantly. The following...

@vmoens , FYI this PR is for working code, not documentation.

@vmoens , I can create a PR there too if you think that has a good likelihood of going through .