JamesOwers midi_degradation_toolkit issues

JamesOwers / midi_degradation_toolkit

A toolkit for generating datasets of midi files which have been degraded to be 'un-musical'.

MIT License

38 stars 5 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

measure_errors doesn't detect any deg parameters

#118 apmcleod opened 4 years ago
2
reading json directly via argparse doesn't work

#117 apmcleod closed 4 years ago
1
Interrupting during ./make_dataset download step results in error

#116 apmcleod closed 4 years ago
3
Add motivation and examples for data augmentation.

#115 JamesOwers closed 4 years ago
1
Maybe different error handling

#114 apmcleod closed 4 years ago
0
data_structures.read_note_csv cannot read our own CSVs

#113 apmcleod closed 4 years ago
1
Some tests failed in data_structures

#112 apmcleod closed 4 years ago
2
Missing Dependency: pytest

#111 apmcleod closed 4 years ago
0
urllib import errors

#110 apmcleod closed 4 years ago
2
Rule

#109 apmcleod closed 4 years ago
0
Just some ideas

#108 benedekrozemberczki closed 4 years ago
2
Rerun Task 2

#107 JamesOwers closed 4 years ago
1
add tests to test_data_structures.py

#106 JamesOwers closed 4 years ago
1
download csv files directly from datasets?

#105 JamesOwers opened 5 years ago
1
Decide on use of Composition class

#104 JamesOwers closed 4 years ago
0
data_structures fix_overlapping_{notes,pitches} catch introduction of dur 0 notes

#103 JamesOwers closed 4 years ago
2
Remove or implement baseline arg in eval_task.py

#102 JamesOwers closed 5 years ago
1
Handle case of zero length notes created in formatters

#101 JamesOwers opened 5 years ago
2
Evaluation in pytorch trainers

#100 JamesOwers opened 5 years ago
1
Implement a load method for loading checkpoints in trainers

#99 JamesOwers opened 5 years ago
0
Trainers and formatters

#98 JamesOwers opened 5 years ago
2
Look into making a readthedocs

#97 JamesOwers closed 4 years ago
0
Make sure we attribute and credit correctly

#96 JamesOwers opened 5 years ago
4
In mdtk.degradations function overlaps, the var offsets is unused

#95 JamesOwers closed 5 years ago
1
Test windows install

#94 apmcleod opened 5 years ago
1
Raise issue with PPDD about empty midi files

#93 JamesOwers opened 5 years ago
0
Remove corpus base path from format_csv_paths

#92 apmcleod closed 5 years ago
3
Add optional flags for pip install such that users can optionally install pytorch etc.

#91 JamesOwers closed 4 years ago
2
Ensure baselines/get_results.py works as a callable script

#90 JamesOwers closed 5 years ago
0
Ensure constant file ordering in make dataset

#89 apmcleod closed 5 years ago
1
Don't suppress all warnings in train and eval

#88 apmcleod closed 4 years ago
3
Probably make --clear default in make_dataset.py

#87 apmcleod closed 5 years ago
0
Org

#86 apmcleod closed 5 years ago
1
Implement --baseline in train and eval scripts

#85 apmcleod opened 5 years ago
3
Update from master

#84 apmcleod closed 5 years ago
0
Decide which scripts, etc to include in the mdtk package

#83 apmcleod closed 4 years ago
2
Add dataloader to degrade on the fly

#82 apmcleod closed 4 years ago
2
Read task 2 labels from degradation_ids.csv

#81 apmcleod closed 5 years ago
0
Support evaluation for people not using our trainers

#80 apmcleod closed 5 years ago
0
Created new get_results script

#79 JamesOwers closed 5 years ago
10
Get results

#78 JamesOwers closed 5 years ago
1
Baseline model improvements and comparisons from literature

#77 JamesOwers opened 5 years ago
1
Change Error Identification -> Error Location throughout.

#76 apmcleod closed 5 years ago
0
double_pianoroll_to_df is very slow.

#75 apmcleod opened 5 years ago
2
Generate docs

#74 apmcleod closed 5 years ago
2
Improvements for training [low priority]

#73 JamesOwers opened 5 years ago
0
Weights

#72 apmcleod closed 5 years ago
0
"Rule-based" baselines

#71 JamesOwers closed 4 years ago
8
Eval script

#70 apmcleod closed 5 years ago
1
Eval

#69 apmcleod closed 5 years ago
2

Previous Next