issues
search
matsengrp
/
vampire
🧛 Deep generative models for TCR sequences 🧛
Apache License 2.0
16
stars
4
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
72 investigate relationship between sequence sharing and likelihood
#73
matsen
closed
5 years ago
0
Investigate relationship between sequence sharing and likelihood
#72
matsen
closed
5 years ago
1
70 file paths
#71
matsen
closed
5 years ago
0
Simplify file paths by joining with outdir
#70
matsen
closed
5 years ago
0
68 generalization performance
#69
matsen
closed
5 years ago
0
Investigate generalization performance
#68
matsen
closed
5 years ago
0
Hierarchical model
#67
matsen
opened
5 years ago
0
54 full implementation + pipeline for thymic Q
#66
matsen
closed
5 years ago
1
63 replicates
#65
matsen
closed
5 years ago
0
54 & 63 progress: Q and scons replicates
#64
matsen
closed
5 years ago
0
Add replicate runs
#63
matsen
closed
5 years ago
2
Batch norm
#62
matsen
closed
5 years ago
1
58 beta schedule
#61
matsen
closed
5 years ago
1
Moving to val_loss for EarlyStopping.
#60
matsen
closed
5 years ago
0
Try using val_loss again now that we're doing more out-of-sample comparison.
#59
matsen
closed
5 years ago
0
adjust beta during training
#58
matsen
closed
5 years ago
2
SConscript refactoring and R plotting
#57
matsen
closed
5 years ago
0
record command line call
#56
matsen
closed
5 years ago
1
48 fooling
#55
matsen
closed
5 years ago
0
OLGA + thymic Q
#54
matsen
closed
5 years ago
5
40 validation
#53
matsen
closed
5 years ago
0
49 gpu
#52
matsen
closed
5 years ago
0
Allow multiple test sets
#51
matsen
closed
5 years ago
1
preprocessing includes the DNA column, but the head step removes it
#50
matsen
closed
5 years ago
1
run on GPUs
#49
matsen
closed
5 years ago
1
More systematic cross-fooling comparison
#48
matsen
closed
5 years ago
1
27 training
#47
matsen
closed
5 years ago
0
45 loss regression
#46
matsen
closed
5 years ago
0
Use regression to find loss weights
#45
matsen
closed
5 years ago
0
16 1 olga
#44
matsen
closed
5 years ago
0
Add more single-repertoire summaries, and synchronize names with divergence code
#42
matsen
closed
5 years ago
1
Compare sumrep distributions
#41
matsen
closed
5 years ago
1
Add comparisons with a validation set
#40
matsen
closed
5 years ago
1
How many blood samples are needed to get a generalizable model?
#39
matsen
closed
5 years ago
0
rename krdav model to basic
#38
matsen
closed
5 years ago
0
Gene name conversion drops sequences; we don't get quite as many sequences as we ask for
#37
matsen
closed
5 years ago
3
Entrypoint for tcr-vae.
#36
matsen
closed
5 years ago
0
proper entry points and scripts
#35
matsen
closed
5 years ago
0
Versioning
#34
matsen
closed
5 years ago
1
Why are we getting so much lower V loss than J loss?
#33
matsen
closed
5 years ago
2
29 loss
#32
matsen
closed
5 years ago
1
26 cluster via sbatch
#31
matsen
closed
5 years ago
0
SCons tidying
#30
matsen
closed
5 years ago
1
Reconsider losses, including KL term
#29
matsen
closed
5 years ago
10
Try some more models
#28
matsen
closed
5 years ago
1
How much training data is needed?
#27
matsen
closed
5 years ago
0
Use cluster via sbatch
#26
matsen
closed
5 years ago
0
23 count germline matches
#25
matsen
closed
5 years ago
1
21 1 length loss cont
#24
matsen
closed
5 years ago
0
Loss counting germline-matching amino acids
#23
matsen
closed
5 years ago
2
Previous
Next