Making a pipeline for Pruning, Quantization and Knowledge Distillation

SforAiDl / KD_Lib

A Pytorch Knowledge Distillation library for benchmarking and extending works in the domains of Knowledge Distillation, Pruning, and Quantization.

https://kd-lib.readthedocs.io/

MIT License

609 stars 58 forks source link

Open Het-Shah opened 3 years ago

Het-Shah commented 3 years ago

Currently, the user has to import and run all three things independently. Having a pipeline will make the entire work streamlined for the end-user.

The user should be able to add pipelines like [KD, Pruning] or [KD, Quantization].

We can discuss how we want to go about doing this.