issues
search
alanngnet
/
CoverHunterMPS
Fork of Liu Feng's CoverHunter to run on a single computer, plus more features and documentation.
9
stars
4
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
does anyone have a link to the CoverHunter256 checkpoint?
#40
brresnic
opened
1 day ago
2
Improve recognition for performances using non-440Hz tuning
#37
alanngnet
opened
2 months ago
1
Improve recognition for transposed performances
#36
alanngnet
opened
2 months ago
0
Memory leak during train_tune runs
#34
alanngnet
opened
5 months ago
0
can we improve the pitch augmentation method?
#33
alanngnet
closed
6 months ago
1
Memory leak during CQT generation
#32
alanngnet
closed
3 months ago
2
Support arbitrary frequency range and resolution
#31
alanngnet
closed
6 months ago
2
analyze / fix output reliability
#29
alanngnet
opened
7 months ago
8
support arbitrary CQT hop lengths
#28
alanngnet
opened
7 months ago
0
decide on license to use for training and testset data
#27
alanngnet
opened
7 months ago
0
add Mean Rank 1 (rank1) metric to tensorboard viz
#26
alanngnet
opened
7 months ago
0
rename "egs" folder to "training"
#25
alanngnet
closed
7 months ago
0
can training be further performance-optimized?
#24
alanngnet
opened
7 months ago
1
document query_in_ref usage
#23
alanngnet
opened
7 months ago
0
Unit tests
#22
samuel-gauthier
closed
7 months ago
0
add Irish trad sample dataset
#21
alanngnet
closed
7 months ago
2
bug? "too many open files" during long training runs
#20
alanngnet
closed
7 months ago
7
Code cleanup
#19
samuel-gauthier
closed
7 months ago
0
tools: use pytorch/nnAudio for data preparation's CQT
#18
samuel-gauthier
closed
7 months ago
1
can this model be used to do multi-label musical classification, not just CSI?
#17
alanngnet
opened
8 months ago
0
how to incrementally train model with additional tunes/pieces/songs?
#16
alanngnet
opened
8 months ago
1
Use black and ruff to reformat
#15
samuel-gauthier
closed
8 months ago
0
connect `extract_csi_features` so `_cluster_plot()` can have awareness of exactly which labels and which samples were excluded from the training data and mark those in the plot.
#13
alanngnet
closed
8 months ago
0
test suite
#12
alanngnet
opened
8 months ago
1
add AMD GPU support
#11
alanngnet
opened
8 months ago
0
Windows instructions for covers80 data prep
#10
alanngnet
opened
8 months ago
0
need sample training data for specific non-pop musical cultures
#9
alanngnet
opened
8 months ago
1
t-SNE plot improvements?
#8
alanngnet
opened
8 months ago
0
add train/val split to extract_csi_features, and make ratios configurable as external hyperparameters
#7
alanngnet
closed
8 months ago
1
add k-fold technique to dynamically split train/val data to cover all data for production use
#6
alanngnet
closed
6 months ago
5
automate the optimization/tuning process for chunk_frame, spec_augmentation, m_per_class and other culture-specific or data-specific hyperparameters during training
#5
alanngnet
closed
5 months ago
3
add data train/test splitting feature into extract_csi_features script
#4
alanngnet
closed
8 months ago
3
explain purpose/value of separate train, query, and ref datasets
#3
alanngnet
closed
8 months ago
2
build inference script
#2
alanngnet
closed
7 months ago
3
implement coarse-to-fine alignment
#1
alanngnet
opened
8 months ago
1