Code

Setup_audio#322

Code

Merged

monke6942021 merged 65 commits intomainmlcommons/croissant:mainfrom setup_audiomlcommons/croissant:setup_audioCopy head branch name to clipboardFeb 8, 2024

+196-1

Commits

Commits on Sep 14, 2023

• This should make the validation process easier.

  monke6942021

  committed
  ade26eb

  Copy full SHA for ade26eb

• Validation will be easier now.

  monke6942021

  committed
  5faaf02

  Copy full SHA for 5faaf02

• Merge branch 'main' of https://github.com/mlcommons/croissant

  monke6942021

  committed
  b84be69

  Copy full SHA for b84be69

Commits on Nov 8, 2023

• This will let croissant recognize AudioObjects

  monke6942021

  committed
  82b0217

  Copy full SHA for 82b0217

Commits on Nov 16, 2023

• mini_CC_BY_SA dataset has been added

  monke6942021

  committed
  506bd0d

  Copy full SHA for 506bd0d

• This should make the validation process easier.

  monke6942021

  committed
  9861395

  Copy full SHA for 9861395

• mini_CC_BY_SA dataset has been added

  monke6942021

  committed
  50d78ac

  Copy full SHA for 50d78ac

• mini_CC_BY_SA is now part of the non-hermetic tests

  monke6942021

  committed
  369621d

  Copy full SHA for 369621d

• Now it takes input from a Google Drive

  monke6942021

  committed
  5723b3f

  Copy full SHA for 5723b3f

Commits on Nov 20, 2023

• Many of the tests were failing because of librosa, so anything with it has been commented out for now.

  monke6942021

  committed
  4b7150e

  Copy full SHA for 4b7150e

• Hopefully this passes the pr test

  monke6942021

  committed
  fafe57d

  Copy full SHA for fafe57d

• The validation script CLAIMS that it works.

  monke6942021

  committed
  30d2aca

  Copy full SHA for 30d2aca

• Hopefully black[jupyter] helps

  monke6942021

  committed
  5f23744

  Copy full SHA for 5f23744

• Now jupyter has been added before the original black. Let's see if this helps.

  monke6942021

  committed
  56a86e8

  Copy full SHA for 56a86e8

• I can try resetting to see if the formatting test works.

  monke6942021

  committed
  babbbc2

  Copy full SHA for babbbc2

Commits on Dec 7, 2023

• Changed treatment of AudioObject in field.py

  monke6942021

  committed
  8cd0e89

  Copy full SHA for 8cd0e89

Commits on Dec 18, 2023

• This version can load audio in field.py

  monke6942021

  committed
  7d22310

  Copy full SHA for 7d22310

• Merge branch 'main' into setup_audio

  monke6942021

  committed
  363bac6

  Copy full SHA for 363bac6

• Audio files have some level of usability now.

  monke6942021

  committed
  4e5f54e

  Copy full SHA for 4e5f54e

• The exact dataset I worked with

  monke6942021

  committed
  081c129

  Copy full SHA for 081c129

• I got the audio support to work.

  monke6942021

  committed
  5a88008

  Copy full SHA for 5a88008

Commits on Dec 19, 2023

• This should be a good touch up to help with jpeg files too.

  monke6942021

  committed
  154ef14

  Copy full SHA for 154ef14

Commits on Jan 4, 2024

• This is a cheap shot option, but it might pass the unit tests.

  monke6942021

  committed
  28116d8

  Copy full SHA for 28116d8

• Removed URL because it's not being used.

  monke6942021

  committed
  f42626f

  Copy full SHA for f42626f

• librosa is now in ththe correct alphabetical position.

  monke6942021

  committed
  7f94125

  Copy full SHA for 7f94125

• This is a gamble.

  monke6942021

  committed
  7706916

  Copy full SHA for 7706916

• This is another gamble.

  monke6942021

  committed
  7fdc4b8

  Copy full SHA for 7fdc4b8

• Changed import method for librosa.

  monke6942021

  committed
  e4011ea

  Copy full SHA for e4011ea

Commits on Jan 7, 2024

• I changed the pyptoject file. Let's see if it helps.

  monke6942021

  committed
  8be8b76

  Copy full SHA for 8be8b76

• Commented out the librosa import.

  monke6942021

  committed
  dcc8e01

  Copy full SHA for dcc8e01

• Hope this passes PyTest

  monke6942021

  committed
  e9121b2

  Copy full SHA for e9121b2

• Installing black[jupyter]

  monke6942021

  committed
  8924b08

  Copy full SHA for 8924b08

• Storing black[jupyter] as a dependency.

  monke6942021

  committed
  2e24289

  Copy full SHA for 2e24289

• black formatted some files.

  monke6942021

  committed
  5a441e8

  Copy full SHA for 5a441e8

• Another file has been black reformatted

  monke6942021

  committed
  12a0fd2

  Copy full SHA for 12a0fd2

Commits on Jan 10, 2024

• I ran the test locally, and changed the file accordingly. I hope it works.

  monke6942021

  committed
  578127e

  Copy full SHA for 578127e

• I deleted the obvious error.

  monke6942021

  committed
  bec4b4f

  Copy full SHA for bec4b4f

• Changed format of output.

  monke6942021

  committed
  2dd38a7

  Copy full SHA for 2dd38a7

• Changed output to string format.

  monke6942021

  committed
  ffb126d

  Copy full SHA for ffb126d

• I just changed the test case, so this is a huge gamble.

  monke6942021

  committed
  ca6db64

  Copy full SHA for ca6db64

• Pytest wants the data to be in double quotations.

  monke6942021

  committed
  24ffa9f

  Copy full SHA for 24ffa9f

• The output in records.jsonl and the output of the actual function are different, so this is a gamble.

  monke6942021

  committed
  4d48a9a

  Copy full SHA for 4d48a9a

• The output file is going to be very long. I don't see any other way to do it.

  monke6942021

  committed
  6777a07

  Copy full SHA for 6777a07

• I forgot to save the output.

  monke6942021

  committed
  b70229e

  Copy full SHA for b70229e

• Changed the number. Let's see if it works.

  monke6942021

  committed
  5c0eaed

  Copy full SHA for 5c0eaed

• Mentioned that the package name is also librosa.

  monke6942021

  committed
  2a69efd

  Copy full SHA for 2a69efd

• Hope this works.

  monke6942021

  committed
  ab23572

  Copy full SHA for ab23572

• updated dev dependencies.

  monke6942021

  committed
  cf043c8

  Copy full SHA for cf043c8

Commits on Jan 11, 2024

• I used black in a different way, so it might work this time.

  monke6942021

  committed
  fd5393a

  Copy full SHA for fd5393a

Commits on Jan 16, 2024

• addressing the issues that Pierre bought up.

  monke6942021

  committed
  4c40d91

  Copy full SHA for 4c40d91

• Since I used load.py for this one, I'm confident that this should work.

  monke6942021

  committed
  230d1fd

  Copy full SHA for 230d1fd

Commits on Jan 30, 2024

• Changed the name of the library

  monke6942021

  committed
  ba72b11

  Copy full SHA for ba72b11

• removed the commment

  monke6942021

  committed
  0177150

  Copy full SHA for 0177150

• dealt with conflicts.

  monke6942021

  committed
  9f2d083

  Copy full SHA for 9f2d083

• Hopefully I don't need git LFS for this.

  monke6942021

  committed
  c5d38e2

  Copy full SHA for c5d38e2

• black reformatted optional.py

  monke6942021

  committed
  b587614

  Copy full SHA for b587614

• For some reason, my unit test is supposed to be in 0.8

  monke6942021

  committed
  1834c49

  Copy full SHA for 1834c49

Commits on Jan 31, 2024

• Removed audio_test from 0.8.

  monke6942021

  committed
  7611a5a

  Copy full SHA for 7611a5a

• Added fileObject and fileSet in context.

  monke6942021

  committed
  c66524b

  Copy full SHA for c66524b

• I will remove this from 0.8 later, but I need to prove a point.

  monke6942021

  committed
  a777c7f

  Copy full SHA for a777c7f

• Adding audio_test to 0.8 did nothing of value.

  monke6942021

  committed
  8b9b511

  Copy full SHA for 8b9b511

Commits on Feb 8, 2024

• Merge branch 'main' of https://github.com/mlcommons/croissant into setup_audio

  monke6942021

  committed
  b34ede7

  Copy full SHA for b34ede7

• Changed audio_test to be compatible with new changes.

  monke6942021

  committed
  c0ce931

  Copy full SHA for c0ce931

• This should pass the tests, but validation still works a bit differently on my end.

  monke6942021

  committed
  63603e4

  Copy full SHA for 63603e4

• They needed a version of the audio dataset in 0.8

  monke6942021

  committed
  d212703

  Copy full SHA for d212703