Skip to content
GitLab
Menu
Projects
Groups
Snippets
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
Lucas Miranda
deepOF
Commits
bda995ce
Commit
bda995ce
authored
Mar 24, 2021
by
lucas_miranda
Browse files
Rolled back to KNN instead of radius
parent
45a3beea
Pipeline
#96822
canceled with stages
in 4 minutes and 28 seconds
Changes
1
Pipelines
1
Hide whitespace changes
Inline
Side-by-side
deepof_experiments.smk
View file @
bda995ce
...
...
@@ -18,7 +18,7 @@ losses = ["ELBO"] # , "MMD", "ELBO+MMD"]
encodings = [6] # [2, 4, 6, 8, 10, 12, 14, 16]
cluster_numbers = [25] # [1, 5, 10, 15, 20, 25]
latent_reg = ["none", "categorical", "variance", "categorical+variance"]
entropy_
min_
n = [2, 5, 8, 1
2
]
entropy_
kn
n = [2
0
, 5
0
, 8
0
, 1
00
]
pheno_weights = [0.01, 0.1, 0.25, 0.5, 1.0, 2.0, 4.0, 10.0, 100.0]
...
...
@@ -35,12 +35,12 @@ rule deepof_experiments:
# ),
expand(
"/psycl/g/mpsstatgen/lucas/DLC/DLC_autoencoders/DeepOF/deepof/logs/latent_regularization_experiments/trained_weights/"
"GMVAE_loss={loss}_encoding={encs}_k={k}_latreg={latreg}_entropy
mink
={ent
mi
n}_final_weights.h5",
"GMVAE_loss={loss}_encoding={encs}_k={k}_latreg={latreg}_entropy
knn
={ent
kn
n}_final_weights.h5",
loss=losses,
encs=encodings,
k=cluster_numbers,
latreg=latent_reg,
ent
mi
n=entropy_
min_
n,
ent
kn
n=entropy_
kn
n,
),
# expand(
# "/psycl/g/mpsstatgen/lucas/DLC/DLC_autoencoders/DeepOF/deepof/logs/pheno_classification_experiments/trained_weights/"
...
...
@@ -89,7 +89,7 @@ rule latent_regularization_experiments:
output:
trained_models=os.path.join(
outpath,
"latent_regularization_experiments/trained_weights/GMVAE_loss={loss}_encoding={encs}_k={k}_latreg={latreg}_entropy
mink
={ent
mi
n}_final_weights.h5",
"latent_regularization_experiments/trained_weights/GMVAE_loss={loss}_encoding={encs}_k={k}_latreg={latreg}_entropy
knn
={ent
kn
n}_final_weights.h5",
),
shell:
"pipenv run python -m deepof.train_model "
...
...
@@ -106,7 +106,7 @@ rule latent_regularization_experiments:
"--mmd-warmup 20 "
"--montecarlo-kl 10 "
"--encoding-size {wildcards.encs} "
"--entropy-
min-
n {wildcards.ent
mi
n} "
"--entropy-
kn
n {wildcards.ent
kn
n} "
"--batch-size 256 "
"--window-size 24 "
"--window-step 12 "
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment