Coverage for biobb_ml/utils/correlation_matrix.py: 75%

1#!/usr/bin/env python3 

2 

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

4import argparse 

5import pandas as pd 

6import matplotlib.pyplot as plt 

7import seaborn as sns 

8from biobb_common.generic.biobb_object import BiobbObject 

9from biobb_common.configuration import settings 

10from biobb_common.tools import file_utils as fu 

11from biobb_common.tools.file_utils import launchlogger 

12from biobb_ml.utils.common import check_input_path, check_output_path, getHeader, getIndependentVars 

13 

14 

15class CorrelationMatrix(BiobbObject): 

16 """ 

17 | biobb_ml CorrelationMatrix 

18 | Generates a correlation matrix from a given dataset. 

19 

20 Args: 

21 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_correlation_matrix.csv>`_. Accepted formats: csv (edam:format_3752). 

22 output_plot_path (str): Path to the correlation matrix plot. File type: output. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/reference/utils/ref_output_plot_correlation_matrix.png>`_. Accepted formats: png (edam:format_3603). 

23 properties (dic): 

24 * **features** (*dict*) - ({}) Independent variables or columns from your dataset you want to compare. 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. 

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

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

27 

28 Examples: 

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

30 

31 from biobb_ml.utils.correlation_matrix import correlation_matrix 

32 prop = { 

33 'features': { 

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

35 } 

36 } 

37 correlation_matrix(input_dataset_path='/path/to/myDataset.csv', 

38 output_plot_path='/path/to/newPlot.png', 

39 properties=prop) 

40 

41 Info: 

42 * wrapped_software: 

43 * name: In house 

44 * license: Apache-2.0 

45 * ontology: 

46 * name: EDAM 

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

48 

49 """ 

50 

51 def __init__(self, input_dataset_path, output_plot_path, 

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

53 properties = properties or {} 

54 

55 # Call parent class constructor 

56 super().__init__(properties) 

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

58 

59 # Input/Output files 

60 self.io_dict = { 

61 "in": {"input_dataset_path": input_dataset_path}, 

62 "out": {"output_plot_path": output_plot_path} 

63 } 

64 

65 # Properties specific for BB 

66 self.features = properties.get('features', {}) 

67 self.properties = properties 

68 

69 # Check the properties 

70 self.check_properties(properties) 

71 self.check_arguments() 

72 

73 def check_data_params(self, out_log, err_log): 

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

75 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__) 

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

77 

78 @launchlogger 

79 def launch(self) -> int: 

80 """Execute the :class:`CorrelationMatrix <utils.correlation_matrix.CorrelationMatrix>` utils.correlation_matrix.CorrelationMatrix object.""" 

81 

82 # check input/output paths and parameters 

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

84 

85 # Setup Biobb 

86 if self.check_restart(): 

87 return 0 

88 self.stage_files() 

89 

90 # load dataset 

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

92 if 'columns' in self.features: 

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

94 skiprows = 1 

95 else: 

96 labels = None 

97 skiprows = None 

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

99 

100 if self.features: 

101 data = getIndependentVars(self.features, data, self.out_log, self.__class__.__name__) 

102 

103 fu.log('Parsing dataset', self.out_log, self.global_log) 

104 if data.shape[1] < 10: 

105 s = None 

106 fs = 12 

107 elif data.shape[1] >= 10 and data.shape[1] < 20: 

108 s = (12, 12) 

109 fs = 9 

110 elif data.shape[1] >= 20: 

111 s = (16, 16) 

112 fs = 7 

113 

114 f, ax = plt.subplots(figsize=s) 

115 corr = data.corr() 

116 sns.heatmap(round(corr, 2), annot=True, ax=ax, cmap='Blues', fmt='.2f', square=True, annot_kws={"fontsize": fs}) 

117 f.subplots_adjust(top=0.93) 

118 f.suptitle('Attributes Correlation Matrix', fontsize=14) 

119 plt.tight_layout(rect=[0, 0.03, 1, 0.95]) 

120 

121 plt.savefig(self.io_dict["out"]["output_plot_path"], dpi=150) 

122 fu.log('Saving Correlation Matrix Plot to %s' % self.io_dict["out"]["output_plot_path"], self.out_log, self.global_log) 

123 

124 # Copy files to host 

125 self.copy_to_host() 

126 

127 self.tmp_files.extend([ 

128 self.stage_io_dict.get("unique_dir") 

129 ]) 

130 self.remove_tmp_files() 

131 

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

133 

134 return 0 

135 

136 

137def correlation_matrix(input_dataset_path: str, output_plot_path: str, properties: dict = None, **kwargs) -> int: 

138 """Execute the :class:`CorrelationMatrix <utils.correlation_matrix.CorrelationMatrix>` class and 

139 execute the :meth:`launch() <utils.correlation_matrix.CorrelationMatrix.launch>` method.""" 

140 

141 return CorrelationMatrix(input_dataset_path=input_dataset_path, 

142 output_plot_path=output_plot_path, 

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

144 

145 

146def main(): 

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

148 parser = argparse.ArgumentParser(description="Generates a correlation matrix from a given dataset", formatter_class=lambda prog: argparse.RawTextHelpFormatter(prog, width=99999)) 

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

150 

151 # Specific args of each building block 

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

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

154 required_args.add_argument('--output_plot_path', required=True, help='Path to the correlation matrix plot. Accepted formats: png.') 

155 

156 args = parser.parse_args() 

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

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

159 

160 # Specific call of each building block 

161 correlation_matrix(input_dataset_path=args.input_dataset_path, 

162 output_plot_path=args.output_plot_path, 

163 properties=properties) 

164 

165 

166if __name__ == '__main__': 

167 main()