deepof_experiments.smk 4.81 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
# @authors lucasmiranda42
# encoding: utf-8
# deepof_experiments

"""

Snakefile for data and imputation.
Execution: sbatch snakemake
Plot DAG: snakemake --snakefile deepof_experiments.smk --forceall --dag | dot -Tpdf > deepof_experiments_DAG.pdf
Plot rule graph: snakemake --snakefile deepof_experiments.smk --forceall --rulegraph | dot -Tpdf > deepof_experiments_RULEGRAPH.pdf

"""

14
import os
15

lucas_miranda's avatar
lucas_miranda committed
16
outpath = "/psycl/g/mpsstatgen/lucas/DLC/DLC_autoencoders/DeepOF/deepof/logs/"
17
losses = ["ELBO"]  # , "MMD", "ELBO+MMD"]
18
19
20
encodings = [4, 8]  # [2, 4, 6, 8, 10, 12, 14, 16]
cluster_numbers = [25]  # [1, 5, 10, 15, 20, 25]
latent_reg = ["none", "categorical", "variance", "categorical+variance"]
21
pheno_weights = [0.01, 0.1, 0.25, 0.5, 1.0, 2.0, 4.0, 10.0, 100.0]
22

23

24
25
rule deepof_experiments:
    input:
26
27
28
29
30
31
32
33
34
35
36
        # expand( "/psycl/g/mpsstatgen/lucas/DLC/DLC_autoencoders/DeepOF/deepof/logs/hyperparameter_tuning/trained_weights/"
        #         "GMVAE_loss={loss}_encoding=2_run_1_final_weights.h5",
        #         loss=losses,
        # )
        expand(
            "/psycl/g/mpsstatgen/lucas/DLC/DLC_autoencoders/DeepOF/deepof/logs/dimension_and_loss_experiments/trained_weights/"
            "GMVAE_loss={loss}_encoding={encs}_k={k}_latreg={latreg}_final_weights.h5",
            loss=losses,
            encs=encodings,
            k=cluster_numbers,
            latreg=latent_reg,
lucas_miranda's avatar
lucas_miranda committed
37
        )
lucas_miranda's avatar
lucas_miranda committed
38
        # expand(
lucas_miranda's avatar
lucas_miranda committed
39
        #     "/psycl/g/mpsstatgen/lucas/DLC/DLC_autoencoders/DeepOF/deepof/logs/pheno_classification_experiments/trained_weights/"
lucas_miranda's avatar
lucas_miranda committed
40
41
42
43
44
45
        #     "GMVAE_loss={loss}_encoding={encs}_k={k}_pheno={phenos}_run_1_final_weights.h5",
        #     loss=losses,
        #     encs=encodings,
        #     k=cluster_numbers,
        #     phenos=pheno_weights,
        # ),
46
47


48
# rule coarse_hyperparameter_tuning:
lucas_miranda's avatar
lucas_miranda committed
49
#     input:
50
#         data_path="/psycl/g/mpsstatgen/lucas/DLC/DLC_models/deepof_single_topview/",
lucas_miranda's avatar
lucas_miranda committed
51
52
53
#     output:
#         trained_models=os.path.join(
#             outpath,
54
#             "hyperparameter_tuning/trained_weights/GMVAE_loss={loss}_encoding=2_run_1_final_weights.h5",
lucas_miranda's avatar
lucas_miranda committed
55
56
57
58
#         ),
#     shell:
#         "pipenv run python -m deepof.train_model "
#         "--train-path {input.data_path} "
59
60
#         "--val-num 25 "
#         "--components 15 "
lucas_miranda's avatar
lucas_miranda committed
61
62
63
64
65
66
#         "--input-type coords "
#         "--predictor 0 "
#         "--phenotype-classifier 0 "
#         "--variational True "
#         "--loss {wildcards.loss} "
#         "--kl-warmup 20 "
67
68
#         "--mmd-warmup 0 "
#         "--encoding-size 2 "
lucas_miranda's avatar
lucas_miranda committed
69
70
#         "--batch-size 256 "
#         "--window-size 24 "
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
#         "--window-step 12 "
#         "--output-path {outpath}coarse_hyperparameter_tuning "
#         "--hyperparameter-tuning hyperband "
#         "--hpt-trials 3"


rule latent_regularization_experiments:
    input:
        data_path=ancient("/psycl/g/mpsstatgen/lucas/DLC/DLC_models/deepof_single_topview/"),
    output:
        trained_models=os.path.join(
            outpath,
            "latent_regularization_experiments/trained_weights/GMVAE_loss={loss}_encoding={encs}_k={k}_latreg={latreg}_final_weights.h5",
        ),
    shell:
        "pipenv run python -m deepof.train_model "
        "--train-path {input.data_path} "
        "--val-num 5 "
        "--components {wildcards.k} "
        "--input-type coords "
        "--predictor 0 "
        "--phenotype-classifier 0 "
        "--variational True "
        "--latent-reg {wildcards.latreg} "
        "--loss {wildcards.loss} "
        "--kl-warmup 20 "
        "--mmd-warmup 20 "
        "--montecarlo-kl 10 "
        "--encoding-size {wildcards.encs} "
        "--batch-size 256 "
        "--window-size 24 "
        "--window-step 6 "
        "--exclude-bodyparts Tail_base,Tail_1,Tail_2,Tail_tip "
        "--stability-check 3 "
        "--output-path {outpath}latent_regularization_experiments"
lucas_miranda's avatar
lucas_miranda committed
106
107
108
109
#
#
# rule explore_phenotype_classification:
#     input:
lucas_miranda's avatar
lucas_miranda committed
110
#         data_path="/psycl/g/mpsstatgen/lucas/DLC/DLC_models/deepof_single_topview/",
lucas_miranda's avatar
lucas_miranda committed
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
#     output:
#         trained_models=os.path.join(
#             outpath,
#             "pheno_classification_experiments/trained_weights/GMVAE_loss={loss}_encoding={encs}_k={k}_pheno={phenos}_run_1_final_weights.h5",
#         ),
#     shell:
#         "pipenv run python -m deepof.train_model "
#         "--train-path {input.data_path} "
#         "--val-num 15 "
#         "--components {wildcards.k} "
#         "--input-type coords "
#         "--predictor 0 "
#         "--phenotype-classifier {wildcards.phenos} "
#         "--variational True "
#         "--loss {wildcards.loss} "
#         "--kl-warmup 20 "
#         "--mmd-warmup 20 "
#         "--montecarlo-kl 10 "
#         "--encoding-size {wildcards.encs} "
#         "--batch-size 256 "
#         "--window-size 11 "
#         "--window-step 11 "
#         "--stability-check 3  "
#         "--output-path {outpath}pheno_classification_experiments"