1. 09 Sep, 2021 1 commit
  2. 30 Jun, 2021 1 commit
    • Kai Zhang's avatar
      Fix typo in quantization callback · e830629a
      Kai Zhang authored
      Summary: "fb" -> "fn"
      
      Reviewed By: ananthsub
      
      Differential Revision: D29480559
      
      fbshipit-source-id: 78a0cd3ddd25df2c877514d4a5c0c29c248267a2
      e830629a
  3. 26 Jun, 2021 1 commit
    • Kai Zhang's avatar
      Fix quantization test failure · 1894f8a3
      Kai Zhang authored
      Summary:
      # Context
      In post training quantization callback, we make a deepcopy of the Lightning module before validation start and prepare the copy with FX quantization API. The callback keeps the prepared model inside it.
      
      # The problem
      By the second time we run the validation epoch, we try to make a copy of the Lightning module, which has a reference to trainer, which has a reference to quantization callback, which has a prepared model, which is not deepcopiable.
      
      # Mitigation
      Delete the trainer before making a deepcopy.
      We're already doing that in stl/callbacks/quantization, but the changes were not ported into D2 (https://github.com/facebookresearch/d2go/commit/4169abc18ec539a24081b179fcbbc5a5754d102b)Go.
      
      Reviewed By: zhanghang1989
      
      Differential Revision: D29409085
      
      fbshipit-source-id: 24550124181673b2e567b2a04563bcdfb440e145
      1894f8a3
  4. 17 Apr, 2021 2 commits
    • Kai Zhang's avatar
      Delegate to model's customization · aeb24a92
      Kai Zhang authored
      Summary: Delegate FX quantization callback's customization to model.
      
      Reviewed By: wat3rBro
      
      Differential Revision: D27669212
      
      fbshipit-source-id: 2715546cf03134896da6f95ecddaf8503ff95d0b
      aeb24a92
    • Kai Zhang's avatar
      E2E QAT Workflow on Lightning · 845d0b2c
      Kai Zhang authored
      Summary:
      As per title and sanity test E2E QAT workflow on Lightning Trainer.
      
      - add `post_training_opts`. This is required to use `all_steps_qat.json` with Lightning. We don't actually support the post_training_opts in this diff though - we leave it part of T83437359.
      - Update .yaml to specify the Quantize-able modules.
      - Update `lightning_train_net.py` to use the QuantizationAwareTraining callback.
      
      Reviewed By: kandluis
      
      Differential Revision: D26304879
      
      fbshipit-source-id: 948bef4817d385d8a0969e4990d7f17ecd6994b7
      845d0b2c
  5. 31 Mar, 2021 1 commit
  6. 03 Mar, 2021 1 commit
    • Kai Zhang's avatar
      Copy quantization callback to D2go · 5d8068d8
      Kai Zhang authored
      Summary: As titled. Make a copy of quantization callback to unblock D2go OSS.
      
      Reviewed By: zhanghang1989
      
      Differential Revision: D26735525
      
      fbshipit-source-id: 12b77f04cfa1361e856b26ea218a262da1fadd88
      5d8068d8