Skip to content
This repository has been archived by the owner on Nov 1, 2024. It is now read-only.

Rewrite of the load_checkpoint function #650

Open
wants to merge 28 commits into
base: main
Choose a base branch
from
Open

Rewrite of the load_checkpoint function #650

wants to merge 28 commits into from

Commits on Nov 10, 2022

  1. split out a get_checkpoint_path_to_load

    @suchenzang
    suchenzang authored and Peter Albert committed Nov 10, 2022
    Configuration menu
    Copy the full SHA
    e4a3a39 View commit details
    Browse the repository at this point in the history

  2. make restore_file optional, indent in trainer.load_checkpoint path ex… 

    …ists check
    @suchenzang
    suchenzang authored and Peter Albert committed Nov 10, 2022
    Configuration menu
    Copy the full SHA
    1f6071b View commit details
    Browse the repository at this point in the history

  3. switch from default_restore_file string matching to is/not None checks

    @suchenzang
    suchenzang authored and Peter Albert committed Nov 10, 2022
    Configuration menu
    Copy the full SHA
    6ee9c4d View commit details
    Browse the repository at this point in the history

  4. none guard

    @suchenzang
    suchenzang authored and Peter Albert committed Nov 10, 2022
    Configuration menu
    Copy the full SHA
    c68d578 View commit details
    Browse the repository at this point in the history

  5. sorting out restore_file logic

    @suchenzang
    suchenzang authored and Peter Albert committed Nov 10, 2022
    Configuration menu
    Copy the full SHA
    a6fd478 View commit details
    Browse the repository at this point in the history

Commits on Nov 22, 2022

  1. merge main

    Peter Albert committed Nov 22, 2022
    Configuration menu
    Copy the full SHA
    0cf3c0c View commit details
    Browse the repository at this point in the history

Commits on Nov 24, 2022

  1. make save_async a single command

    Peter Albert committed Nov 24, 2022
    Configuration menu
    Copy the full SHA
    18c6d16 View commit details
    Browse the repository at this point in the history

  2. remove duplicate flags

    Peter Albert committed Nov 24, 2022
    Configuration menu
    Copy the full SHA
    570fced View commit details
    Browse the repository at this point in the history

Commits on Nov 25, 2022

  1. change naming to local_checkpoints_dir

    Peter Albert committed Nov 25, 2022
    Configuration menu
    Copy the full SHA
    a9989aa View commit details
    Browse the repository at this point in the history

Commits on Nov 29, 2022

  1. finish core prepare_local_checkpoint_path function

    Peter Albert committed Nov 29, 2022
    Configuration menu
    Copy the full SHA
    6eefb06 View commit details
    Browse the repository at this point in the history

  2. black lint

    Peter Albert committed Nov 29, 2022
    Configuration menu
    Copy the full SHA
    a8b0613 View commit details
    Browse the repository at this point in the history

  3. add dataclass

    Peter Albert committed Nov 29, 2022
    Configuration menu
    Copy the full SHA
    6fe250c View commit details
    Browse the repository at this point in the history

  4. remove old load function, add type hints, move out internal import

    Peter Albert committed Nov 29, 2022
    Configuration menu
    Copy the full SHA
    2ddf1fb View commit details
    Browse the repository at this point in the history

  5. changed naming of prio

    Peter Albert committed Nov 29, 2022
    Configuration menu
    Copy the full SHA
    d46404a View commit details
    Browse the repository at this point in the history

  6. change naming for epoch checkpoints to include num steps

    Peter Albert committed Nov 29, 2022
    Configuration menu
    Copy the full SHA
    725bfb8 View commit details
    Browse the repository at this point in the history

  7. add local caching by including num steps in cache file name

    Peter Albert committed Nov 29, 2022
    Configuration menu
    Copy the full SHA
    5e6218e View commit details
    Browse the repository at this point in the history

Commits on Jan 6, 2023

  1. move checkpoint path out

    Peter Albert committed Jan 6, 2023
    Configuration menu
    Copy the full SHA
    24b4ab8 View commit details
    Browse the repository at this point in the history

Commits on Jan 16, 2023

  1. only merge load_checkpoint related changes

    Peter Albert committed Jan 16, 2023
    Configuration menu
    Copy the full SHA
    cec3941 View commit details
    Browse the repository at this point in the history

Commits on Jan 20, 2023

  1. add types and clean

    Peter Albert committed Jan 20, 2023
    Configuration menu
    Copy the full SHA
    243c327 View commit details
    Browse the repository at this point in the history

Commits on Feb 1, 2023

  1. Merge branch 'main' into peter/rewrite_load_checkpoint

    Peter Albert committed Feb 1, 2023
    Configuration menu
    Copy the full SHA
    d8c0e87 View commit details
    Browse the repository at this point in the history

  2. fixes

    Peter Albert committed Feb 1, 2023
    Configuration menu
    Copy the full SHA
    c3396b8 View commit details
    Browse the repository at this point in the history

Commits on Feb 6, 2023

  1. two paths for local vs nfs checkpoints

    Peter Albert committed Feb 6, 2023
    Configuration menu
    Copy the full SHA
    58dcfb2 View commit details
    Browse the repository at this point in the history

Commits on Feb 10, 2023

  1. add some debugging and load local checkpoints

    Peter Albert committed Feb 10, 2023
    Configuration menu
    Copy the full SHA
    65c03bc View commit details
    Browse the repository at this point in the history

  2. more cleanup, add reset_for_finetuning back in

    Peter Albert committed Feb 10, 2023
    Configuration menu
    Copy the full SHA
    e2efa2d View commit details
    Browse the repository at this point in the history

Commits on Feb 14, 2023

  1. run black .

    Peter Albert committed Feb 14, 2023
    Configuration menu
    Copy the full SHA
    cbff07c View commit details
    Browse the repository at this point in the history

  2. Merge branch 'main' into peter/rewrite_load_checkpoint

    Peter Albert committed Feb 14, 2023
    Configuration menu
    Copy the full SHA
    9366295 View commit details
    Browse the repository at this point in the history

  3. flake8

    Peter Albert committed Feb 14, 2023
    Configuration menu
    Copy the full SHA
    0bde2fe View commit details
    Browse the repository at this point in the history

  4. flake 8

    Peter Albert committed Feb 14, 2023
    Configuration menu
    Copy the full SHA
    7991503 View commit details
    Browse the repository at this point in the history