Coverage for biobb_ml/utils/scale_columns.py: 78%

1#!/usr/bin/env python3 

2 

3"""Module containing the ScaleColumns class and the command line interface.""" 

4import argparse 

5import pandas as pd 

6from biobb_common.generic.biobb_object import BiobbObject 

7from sklearn.preprocessing import MinMaxScaler 

8from biobb_common.configuration import settings 

9from biobb_common.tools import file_utils as fu 

10from biobb_common.tools.file_utils import launchlogger 

11from biobb_ml.utils.common import check_input_path, check_output_path, getHeader, getIndependentVarsList, getTargetsList 

12 

13 

14class ScaleColumns(BiobbObject): 

15 """ 

16 | biobb_ml ScaleColumns 

17 | Scales columns from a given dataset. 

18 

19 Args: 

20 input_dataset_path (str): Path to the input dataset. File type: input. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/data/utils/dataset_scale.csv>`_. Accepted formats: csv (edam:format_3752). 

21 output_dataset_path (str): Path to the output dataset. File type: output. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/reference/utils/ref_output_scale.csv>`_. Accepted formats: csv (edam:format_3752). 

22 properties (dic): 

23 * **targets** (*dict*) - ({}) Independent variables or columns from your dataset you want to scale. You can specify either a list of columns names from your input dataset, a list of columns indexes or a range of columns indexes. Formats: { "columns": ["column1", "column2"] } or { "indexes": [0, 2, 3, 10, 11, 17] } or { "range": [[0, 20], [50, 102]] }. In case of mulitple formats, the first one will be picked. 

24 * **remove_tmp** (*bool*) - (True) [WF property] Remove temporal files. 

25 * **restart** (*bool*) - (False) [WF property] Do not execute if output files exist. 

26 

27 Examples: 

28 This is a use example of how to use the building block from Python:: 

29 

30 from biobb_ml.utils.scale_columns import scale_columns 

31 prop = { 

32 'targets': { 

33 'columns': [ 'column1', 'column2', 'column3' ] 

34 } 

35 } 

36 scale_columns(input_dataset_path='/path/to/myDataset.csv', 

37 output_dataset_path='/path/to/newDataset.csv', 

38 properties=prop) 

39 

40 Info: 

41 * wrapped_software: 

42 * name: In house 

43 * license: Apache-2.0 

44 * ontology: 

45 * name: EDAM 

46 * schema: http://edamontology.org/EDAM.owl 

47 

48 """ 

49 

50 def __init__(self, input_dataset_path, output_dataset_path, 

51 properties=None, **kwargs) -> None: 

52 properties = properties or {} 

53 

54 # Call parent class constructor 

55 super().__init__(properties) 

56 self.locals_var_dict = locals().copy() 

57 

58 # Input/Output files 

59 self.io_dict = { 

60 "in": {"input_dataset_path": input_dataset_path}, 

61 "out": {"output_dataset_path": output_dataset_path} 

62 } 

63 

64 # Properties specific for BB 

65 self.targets = properties.get('targets', {}) 

66 self.properties = properties 

67 

68 # Check the properties 

69 self.check_properties(properties) 

70 self.check_arguments() 

71 

72 def check_data_params(self, out_log, err_log): 

73 """ Checks all the input/output paths and parameters """ 

74 self.io_dict["in"]["input_dataset_path"] = check_input_path(self.io_dict["in"]["input_dataset_path"], "input_dataset_path", out_log, self.__class__.__name__) 

75 self.io_dict["out"]["output_dataset_path"] = check_output_path(self.io_dict["out"]["output_dataset_path"], "output_dataset_path", False, out_log, self.__class__.__name__) 

76 

77 @launchlogger 

78 def launch(self) -> int: 

79 """Execute the :class:`ScaleColumns <utils.scale_columns.ScaleColumns>` utils.scale_columns.ScaleColumns object.""" 

80 

81 # check input/output paths and parameters 

82 self.check_data_params(self.out_log, self.err_log) 

83 

84 # Setup Biobb 

85 if self.check_restart(): 

86 return 0 

87 self.stage_files() 

88 

89 # load dataset 

90 fu.log('Getting dataset from %s' % self.io_dict["in"]["input_dataset_path"], self.out_log, self.global_log) 

91 if 'columns' in self.targets: 

92 labels = getHeader(self.io_dict["in"]["input_dataset_path"]) 

93 skiprows = 1 

94 header = 0 

95 else: 

96 labels = None 

97 skiprows = None 

98 header = None 

99 data = pd.read_csv(self.io_dict["in"]["input_dataset_path"], header=None, sep="\\s+|;|:|,|\t", engine="python", skiprows=skiprows, names=labels) 

100 

101 targets = getTargetsList(self.targets, 'scale', self.out_log, self.__class__.__name__) 

102 

103 fu.log('Scaling [%s] columns from dataset' % getIndependentVarsList(self.targets), self.out_log, self.global_log) 

104 if not self.targets: 

105 df_scaled = data 

106 else: 

107 df_scaled = (data[targets]) 

108 

109 scaler = MinMaxScaler() 

110 

111 df_scaled = pd.DataFrame(scaler.fit_transform(df_scaled)) 

112 

113 data[targets] = df_scaled 

114 

115 hdr = False 

116 if header == 0: 

117 hdr = True 

118 fu.log('Saving dataset to %s' % self.io_dict["out"]["output_dataset_path"], self.out_log, self.global_log) 

119 data.to_csv(self.io_dict["out"]["output_dataset_path"], index=False, header=hdr) 

120 

121 # Copy files to host 

122 self.copy_to_host() 

123 

124 self.tmp_files.extend([ 

125 self.stage_io_dict.get("unique_dir") 

126 ]) 

127 self.remove_tmp_files() 

128 

129 self.check_arguments(output_files_created=True, raise_exception=False) 

130 

131 return 0 

132 

133 

134def scale_columns(input_dataset_path: str, output_dataset_path: str, properties: dict = None, **kwargs) -> int: 

135 """Execute the :class:`ScaleColumns <utils.scale_columns.ScaleColumns>` class and 

136 execute the :meth:`launch() <utils.scale_columns.ScaleColumns.launch>` method.""" 

137 

138 return ScaleColumns(input_dataset_path=input_dataset_path, 

139 output_dataset_path=output_dataset_path, 

140 properties=properties, **kwargs).launch() 

141 

142 

143def main(): 

144 """Command line execution of this building block. Please check the command line documentation.""" 

145 parser = argparse.ArgumentParser(description="Scales columns from a given dataset", formatter_class=lambda prog: argparse.RawTextHelpFormatter(prog, width=99999)) 

146 parser.add_argument('--config', required=False, help='Configuration file') 

147 

148 # Specific args of each building block 

149 required_args = parser.add_argument_group('required arguments') 

150 required_args.add_argument('--input_dataset_path', required=True, help='Path to the input dataset. Accepted formats: csv.') 

151 required_args.add_argument('--output_dataset_path', required=True, help='Path to the output dataset. Accepted formats: csv.') 

152 

153 args = parser.parse_args() 

154 args.config = args.config or "{}" 

155 properties = settings.ConfReader(config=args.config).get_prop_dic() 

156 

157 # Specific call of each building block 

158 scale_columns(input_dataset_path=args.input_dataset_path, 

159 output_dataset_path=args.output_dataset_path, 

160 properties=properties) 

161 

162 

163if __name__ == '__main__': 

164 main()