deepof_experiments.smk 4.95 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
encodings = [6]  # [2, 4, 6, 8, 10, 12, 14, 16]
19
20
cluster_numbers = [25]  # [1, 5, 10, 15, 20, 25]
latent_reg = ["none", "categorical", "variance", "categorical+variance"]
21
entropy_knn = [20, 50, 80, 100]
22
pheno_weights = [0.01, 0.1, 0.25, 0.5, 1.0, 2.0, 4.0, 10.0, 100.0]
23

24

25
26
rule deepof_experiments:
    input:
27
28
29
30
31
32
33
34
35
        # expand(
        #     os.path.join(
        #         outpath,
        #         "coarse_hyperparameter_tuning/trained_weights/GMVAE_loss={loss}_k={k}_encoding={enc}_final_weights.h5",
        #     ),
        #     loss=losses,
        #     k=cluster_numbers,
        #     enc=encodings,
        # ),
Lucas Miranda's avatar
Lucas Miranda committed
36
        expand(
37
            "/psycl/g/mpsstatgen/lucas/DLC/DLC_autoencoders/DeepOF/deepof/logs/latent_regularization_experiments/trained_weights/"
38
            "GMVAE_loss={loss}_encoding={encs}_k={k}_latreg={latreg}_entropyknn={entknn}_final_weights.h5",
39
40
41
42
            loss=losses,
            encs=encodings,
            k=cluster_numbers,
            latreg=latent_reg,
43
            entknn=entropy_knn,
44
45
46
47
48
49
50
51
52
        ),
        # expand(
        #     "/psycl/g/mpsstatgen/lucas/DLC/DLC_autoencoders/DeepOF/deepof/logs/pheno_classification_experiments/trained_weights/"
        #     "GMVAE_loss={loss}_encoding={encs}_k={k}_pheno={phenos}_run_1_final_weights.h5",
        #     loss=losses,
        #     encs=encodings,
        #     k=cluster_numbers,
        #     phenos=pheno_weights,
        # ),
53
54


55
rule coarse_hyperparameter_tuning:
56
    input:
57
        data_path="/psycl/g/mpsstatgen/lucas/DLC/DLC_models/deepof_single_topview/",
58
59
60
    output:
        trained_models=os.path.join(
            outpath,
61
            "coarse_hyperparameter_tuning/trained_weights/GMVAE_loss={loss}_k={k}_encoding={enc}_final_weights.h5",
62
63
64
65
        ),
    shell:
        "pipenv run python -m deepof.train_model "
        "--train-path {input.data_path} "
66
        "--val-num 25 "
67
        "--components {wildcards.k} "
68
69
70
71
72
73
        "--input-type coords "
        "--predictor 0 "
        "--phenotype-classifier 0 "
        "--variational True "
        "--loss {wildcards.loss} "
        "--kl-warmup 20 "
74
        "--mmd-warmup 0 "
75
        "--encoding-size {wildcards.enc} "
76
77
        "--batch-size 256 "
        "--window-size 24 "
78
        "--window-step 12 "
79
80
        "--output-path {outpath}coarse_hyperparameter_tuning "
        "--hyperparameter-tuning hyperband "
81
        "--hpt-trials 1"
82
83


84
85
rule latent_regularization_experiments:
    input:
86
87
88
        data_path=ancient(
            "/psycl/g/mpsstatgen/lucas/DLC/DLC_models/deepof_single_topview/"
        ),
89
90
91
    output:
        trained_models=os.path.join(
            outpath,
92
            "latent_regularization_experiments/trained_weights/GMVAE_loss={loss}_encoding={encs}_k={k}_latreg={latreg}_entropyknn={entknn}_final_weights.h5",
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
        ),
    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} "
109
        "--entropy-knn {wildcards.entknn} "
110
111
112
        "--batch-size 256 "
        "--window-size 24 "
        "--window-step 12 "
113

114
        "--output-path {outpath}latent_regularization_experiments"
115
        # "--exclude-bodyparts Tail_base,Tail_1,Tail_2,Tail_tip "
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144


rule explore_phenotype_classification:
    input:
        data_path="/psycl/g/mpsstatgen/lucas/DLC/DLC_models/deepof_single_topview/",
    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"