mesh

mesh copied to clipboard

Mesh-tf model conversion to onnx?

2

Hi, I've been trying to deploy an mtf model to the NVIDIA Triton Inference Server by converting the SavedModel (output of model.export()) to an onnx file with no luck. I've...

b-analyst

Hi , Thanks for your great work. I am reading your implementation of switch transformers(MoE), I am wondering if you have released the code for finetuning the MoE models on...

fym0503

Internal

1

Internal

copybara-service[bot]

Automated refactoring to make code Python 3 compatible.

1

Automated refactoring to make code Python 3 compatible.

copybara-service[bot]

Internal change

1

Internal change

copybara-service[bot]

Fixing model export breakage.

1

Fixing model export breakage.

copybara-service[bot]

Decode Unicode strings in inference mode.

1

Decode Unicode strings in inference mode. This is already done in eval mode. This CL applies the same logic for inference.

copybara-service[bot]

internal

1

internal

copybara-service[bot]

Use multiple target objectives for distillation. Also see cl/356382304

2

Use multiple target objectives for distillation. Also see cl/356382304

copybara-service[bot]

Add loss functions for multiple-target objectives for distillation.

2

Add loss functions for multiple-target objectives for distillation.

copybara-service[bot]