Natooz/MidiTok

Issues

• KeyError in collators.py line 164

  #198 opened 11 days ago by kroll-software
  1

• Trailing Bars are not appended into REMI token sequence if last note is multibar length

  #200 opened 6 days ago by Mintas
  1

• split_files_for_training() throws error: pyvec::index out of range: -1

  #195 opened 9 days ago by kroll-software
  5

• KeyError in midi_tokenizer.py line 1712

  #197 opened 11 days ago by kroll-software
  3

• Using pitch bend in Octuple tokenizer

  #196 opened 21 days ago by Neptune-S-777
  2

• Environment Issue (Nanobind Error)

  #190 opened a month ago by MHZ9825
  6

• REMI with rests does not produce Bar tokens if rest is at the end of a bar

  #189 opened 2 months ago by mister-magpie
  4

• Error running the 2nd example in readme

  #157 opened 6 months ago by MikeMpapa
  12

• after tokenizing with trained tokenizer, the "tokens" array contains original tokens

  #166 opened 2 months ago by theglassofwater
  5

• Dration token meaning in REMI

  #174 opened 2 months ago by PRamoneda
  5

• Structured tokenizer order is different from documentation

  #169 opened 3 months ago by mister-magpie
  3

• How to choose the vocab_size for training a bpe tokenizer?

  #179 opened 3 months ago by paranoid2droid
  4

• Defining Custom Tokens

  #171 opened 3 months ago by MikeMpapa
  3

• Cannot install it

  #178 opened 4 months ago by shawn120
  6

• Handling Incompatible Files During `split_files_for_training`

  #176 opened 4 months ago by Kinyugo
  20

• Error in training tokenizer "UnicodeDecodeError: 'utf-8' codec can't decode byte 0xf0 in position 0: invalid continuation byte"

  #159 opened 4 months ago by ojus1
  16

• The Example_HuggingFace_Mistral_Transformer.ipynb notebook gives a ValueError when running trainer.train()

  #163 opened 4 months ago by briane412
  5

• Installation Error with miditok

  #155 opened 5 months ago by LiuZH-19
  14

• Add the possibility to train tokenizer with other token aggregation techniques (WordPiece, Unigram...)

  #154 opened 5 months ago by Natooz
  2

• Time signature type issue when using from_dict()

  #151 opened 6 months ago by JLenzy
  4

• Slow Performance of `tokenize_midi_dataset` Function

  #147 opened 6 months ago by Kinyugo
  31

• cannot import DatasetTok - Error with version 3.0.2

  #158 opened 6 months ago by seb-son
  3

• ERROR:model.generate(input_tokens.ids, max_length=200)-------AttributeError: 'list' object has no attribute 'ids'

  #86 opened 8 months ago by geng-lee
  8

• KeyError: 'Chord_ukn4'

  #77 opened 8 months ago by VDT5702
  15

• Data Preparation Issues

  #126 opened 8 months ago by laceyp99
  4

• Special token question

  #140 opened 8 months ago by oiabtt
  1

• Unable to apply BPE or convert BPE-encoded JSON to MIDI

  #137 opened 8 months ago by oiabtt
  11

• Docs and code bug

  #135 opened 8 months ago by oiabtt
  2

• dump bug

  #134 opened 8 months ago by oiabtt
  0

• Support huggingface AutoTokenizer

  #127 opened 8 months ago by oiabtt
  5

• [Enhancement] Versionning documentation

  #133 opened 8 months ago by leleogere
  4

• Need Help! I run the Full_Example_HuggingFace_GPT2_Transformer.ipynb but the output is none

  #128 opened 8 months ago by sunyrain
  4

• Add a faster midi parsing backend

  #112 opened 8 months ago by Yikai-Liao
  59

• Incorrect Placement of Time Signatures

  #131 opened 8 months ago by EterDelta
  1

• Parallelizing tokenization? (enhancement)

  #102 opened 9 months ago by drscotthawley
  6

• Raise IndexError in Tokenizer

  #104 opened 9 months ago by feiyuehchen
  11

• Windows: FileExistsError running data_augmentation on data_set.

  #111 opened 9 months ago by efraimdahl
  3

• What exactly is beat_res?

  #117 opened 9 months ago by parneyw
  3

• `tokens_to_midi` returning a MidiFile that has no notes and one instrument

  #118 opened 9 months ago by parneyw
  2

• Feature: Adding contributors section to the README.md file.

  #119 opened 9 months ago by kalyani2003
  4

• Duration augmentation not implemented in `data_augmentation_midi`

  #85 opened 10 months ago by leleogere
  8

• FileExists error when augmenting midi dataset with tokenize_midi_dataset

  #107 opened 10 months ago by parneyw
  4

• Raise error or warn if same file name in `tokenize_midi_dataset`

  #79 opened 10 months ago by leleogere
  4

• Add ability to add custom labels to DatasetTok

  #78 opened 10 months ago by leleogere
  7

• When using REMI for tokenization with use_time_signatures=True, many duplicate measures can be encoded.

  #74 opened 10 months ago by Chunyuan-Li
  24

• List index out of range when using tokenizer.learn_bpe

  #90 opened a year ago by VladPetk
  5

• Please check your two notebook demos, there are too many errors inside.

  #88 opened a year ago by oiabtt
  1

• Design of program tokens in MIDILike tokenization

  #71 opened a year ago by caenopy
  4

• MIDILike needs note events sorted

  #69 opened a year ago by caenopy
  3

• time_signature_changes mismatch when using REMI tokenizer

  #73 opened a year ago by Chunyuan-Li
  2