comparison pca.py @ 19:d67dcd63f6cb draft

"planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/sklearn commit e2a5eade6d0e5ddf3a47630381a0ad90d80e8a04"
author bgruening
date Tue, 13 Apr 2021 17:32:55 +0000
parents 02d2be90dedb
children
comparison
equal deleted inserted replaced
18:02d2be90dedb 19:d67dcd63f6cb
1 import argparse 1 import argparse
2
2 import numpy as np 3 import numpy as np
3 from sklearn.decomposition import PCA, IncrementalPCA, KernelPCA
4 from galaxy_ml.utils import read_columns 4 from galaxy_ml.utils import read_columns
5 from sklearn.decomposition import IncrementalPCA, KernelPCA, PCA
6
5 7
6 def main(): 8 def main():
7 parser = argparse.ArgumentParser(description='RDKit screen') 9 parser = argparse.ArgumentParser(description="RDKit screen")
8 parser.add_argument('-i', '--infile', 10 parser.add_argument("-i", "--infile", help="Input file")
9 help="Input file") 11 parser.add_argument(
10 parser.add_argument('--header', action='store_true', help="Include the header row or skip it") 12 "--header", action="store_true", help="Include the header row or skip it"
11 parser.add_argument('-c', '--columns', type=str.lower, default='all', choices=['by_index_number', 'all_but_by_index_number',\ 13 )
12 'by_header_name', 'all_but_by_header_name', 'all_columns'], 14 parser.add_argument(
13 help="Choose to select all columns, or exclude/include some") 15 "-c",
14 parser.add_argument('-ci', '--column_indices', type=str.lower, 16 "--columns",
15 help="Choose to select all columns, or exclude/include some") 17 type=str.lower,
16 parser.add_argument('-n', '--number', nargs='?', type=int, default=None,\ 18 default="all",
17 help="Number of components to keep. If not set, all components are kept") 19 choices=[
18 parser.add_argument('--whiten', action='store_true', help="Whiten the components") 20 "by_index_number",
19 parser.add_argument('-t', '--pca_type', type=str.lower, default='classical', choices=['classical', 'incremental', 'kernel'], 21 "all_but_by_index_number",
20 help="Choose which flavour of PCA to use") 22 "by_header_name",
21 parser.add_argument('-s', '--svd_solver', type=str.lower, default='auto', choices=['auto', 'full', 'arpack', 'randomized'], 23 "all_but_by_header_name",
22 help="Choose the type of svd solver.") 24 "all_columns",
23 parser.add_argument('-b', '--batch_size', nargs='?', type=int, default=None,\ 25 ],
24 help="The number of samples to use for each batch") 26 help="Choose to select all columns, or exclude/include some",
25 parser.add_argument('-k', '--kernel', type=str.lower, default='linear',\ 27 )
26 choices=['linear', 'poly', 'rbf', 'sigmoid', 'cosine', 'precomputed'], 28 parser.add_argument(
27 help="Choose the type of kernel.") 29 "-ci",
28 parser.add_argument('-g', '--gamma', nargs='?', type=float, default=None, 30 "--column_indices",
29 help='Kernel coefficient for rbf, poly and sigmoid kernels. Ignored by other kernels') 31 type=str.lower,
30 parser.add_argument('-tol', '--tolerance', type=float, default=0.0, 32 help="Choose to select all columns, or exclude/include some",
31 help='Convergence tolerance for arpack. If 0, optimal value will be chosen by arpack') 33 )
32 parser.add_argument('-mi', '--max_iter', nargs='?', type=int, default=None,\ 34 parser.add_argument(
33 help="Maximum number of iterations for arpack") 35 "-n",
34 parser.add_argument('-d', '--degree', type=int, default=3,\ 36 "--number",
35 help="Degree for poly kernels. Ignored by other kernels") 37 nargs="?",
36 parser.add_argument('-cf', '--coef0', type=float, default=1.0, 38 type=int,
37 help='Independent term in poly and sigmoid kernels') 39 default=None,
38 parser.add_argument('-e', '--eigen_solver', type=str.lower, default='auto', choices=['auto', 'dense', 'arpack'], 40 help="Number of components to keep. If not set, all components are kept",
39 help="Choose the type of eigen solver.") 41 )
40 parser.add_argument('-o', '--outfile', 42 parser.add_argument("--whiten", action="store_true", help="Whiten the components")
41 help="Base name for output file (no extension).") 43 parser.add_argument(
44 "-t",
45 "--pca_type",
46 type=str.lower,
47 default="classical",
48 choices=["classical", "incremental", "kernel"],
49 help="Choose which flavour of PCA to use",
50 )
51 parser.add_argument(
52 "-s",
53 "--svd_solver",
54 type=str.lower,
55 default="auto",
56 choices=["auto", "full", "arpack", "randomized"],
57 help="Choose the type of svd solver.",
58 )
59 parser.add_argument(
60 "-b",
61 "--batch_size",
62 nargs="?",
63 type=int,
64 default=None,
65 help="The number of samples to use for each batch",
66 )
67 parser.add_argument(
68 "-k",
69 "--kernel",
70 type=str.lower,
71 default="linear",
72 choices=["linear", "poly", "rbf", "sigmoid", "cosine", "precomputed"],
73 help="Choose the type of kernel.",
74 )
75 parser.add_argument(
76 "-g",
77 "--gamma",
78 nargs="?",
79 type=float,
80 default=None,
81 help="Kernel coefficient for rbf, poly and sigmoid kernels. Ignored by other kernels",
82 )
83 parser.add_argument(
84 "-tol",
85 "--tolerance",
86 type=float,
87 default=0.0,
88 help="Convergence tolerance for arpack. If 0, optimal value will be chosen by arpack",
89 )
90 parser.add_argument(
91 "-mi",
92 "--max_iter",
93 nargs="?",
94 type=int,
95 default=None,
96 help="Maximum number of iterations for arpack",
97 )
98 parser.add_argument(
99 "-d",
100 "--degree",
101 type=int,
102 default=3,
103 help="Degree for poly kernels. Ignored by other kernels",
104 )
105 parser.add_argument(
106 "-cf",
107 "--coef0",
108 type=float,
109 default=1.0,
110 help="Independent term in poly and sigmoid kernels",
111 )
112 parser.add_argument(
113 "-e",
114 "--eigen_solver",
115 type=str.lower,
116 default="auto",
117 choices=["auto", "dense", "arpack"],
118 help="Choose the type of eigen solver.",
119 )
120 parser.add_argument(
121 "-o", "--outfile", help="Base name for output file (no extension)."
122 )
42 args = parser.parse_args() 123 args = parser.parse_args()
43 124
44 usecols = None 125 usecols = None
45 cols = []
46 pca_params = {} 126 pca_params = {}
47 127
48 if args.columns == 'by_index_number' or args.columns == 'all_but_by_index_number': 128 if args.columns == "by_index_number" or args.columns == "all_but_by_index_number":
49 usecols = [int(i) for i in args.column_indices.split(',')] 129 usecols = [int(i) for i in args.column_indices.split(",")]
50 elif args.columns == 'by_header_name' or args.columns == 'all_but_by_header_name': 130 elif args.columns == "by_header_name" or args.columns == "all_but_by_header_name":
51 usecols = args.column_indices 131 usecols = args.column_indices
52 132
53 header = 'infer' if args.header else None 133 header = "infer" if args.header else None
54 134
55 pca_input = read_columns( 135 pca_input = read_columns(
56 f=args.infile, 136 f=args.infile,
57 c=usecols, 137 c=usecols,
58 c_option=args.columns, 138 c_option=args.columns,
59 sep='\t', 139 sep="\t",
60 header=header, 140 header=header,
61 parse_dates=True, 141 parse_dates=True,
62 encoding=None, 142 encoding=None,
63 index_col=None) 143 index_col=None,
144 )
64 145
65 pca_params.update({'n_components': args.number}) 146 pca_params.update({"n_components": args.number})
66 147
67 if args.pca_type == 'classical': 148 if args.pca_type == "classical":
68 pca_params.update({'svd_solver': args.svd_solver, 'whiten': args.whiten}) 149 pca_params.update({"svd_solver": args.svd_solver, "whiten": args.whiten})
69 if args.svd_solver == 'arpack': 150 if args.svd_solver == "arpack":
70 pca_params.update({'tol': args.tolerance}) 151 pca_params.update({"tol": args.tolerance})
71 pca = PCA() 152 pca = PCA()
72 153
73 elif args.pca_type == 'incremental': 154 elif args.pca_type == "incremental":
74 pca_params.update({'batch_size': args.batch_size, 'whiten': args.whiten}) 155 pca_params.update({"batch_size": args.batch_size, "whiten": args.whiten})
75 pca = IncrementalPCA() 156 pca = IncrementalPCA()
76 157
77 elif args.pca_type == 'kernel': 158 elif args.pca_type == "kernel":
78 pca_params.update({'kernel': args.kernel, 'eigen_solver': args.eigen_solver, 'gamma': args.gamma}) 159 pca_params.update(
160 {
161 "kernel": args.kernel,
162 "eigen_solver": args.eigen_solver,
163 "gamma": args.gamma,
164 }
165 )
79 166
80 if args.kernel == 'poly': 167 if args.kernel == "poly":
81 pca_params.update({'degree': args.degree, 'coef0': args.coef0}) 168 pca_params.update({"degree": args.degree, "coef0": args.coef0})
82 elif args.kernel == 'sigmoid': 169 elif args.kernel == "sigmoid":
83 pca_params.update({'coef0': args.coef0}) 170 pca_params.update({"coef0": args.coef0})
84 elif args.kernel == 'precomputed': 171 elif args.kernel == "precomputed":
85 pca_input = np.dot(pca_input, pca_input.T) 172 pca_input = np.dot(pca_input, pca_input.T)
86 173
87 if args.eigen_solver == 'arpack': 174 if args.eigen_solver == "arpack":
88 pca_params.update({'tol': args.tolerance, 'max_iter': args.max_iter}) 175 pca_params.update({"tol": args.tolerance, "max_iter": args.max_iter})
89 176
90 pca = KernelPCA() 177 pca = KernelPCA()
91 178
92 print(pca_params) 179 print(pca_params)
93 pca.set_params(**pca_params) 180 pca.set_params(**pca_params)
94 pca_output = pca.fit_transform(pca_input) 181 pca_output = pca.fit_transform(pca_input)
95 np.savetxt(fname=args.outfile, X=pca_output, fmt='%.4f', delimiter='\t') 182 np.savetxt(fname=args.outfile, X=pca_output, fmt="%.4f", delimiter="\t")
96 183
97 184
98 if __name__ == "__main__": 185 if __name__ == "__main__":
99 main() 186 main()