Coverage for biobb_ml/regression/regression_predict.py: 75%
97 statements
« prev ^ index » next coverage.py v7.5.1, created at 2024-05-07 09:39 +0000
1#!/usr/bin/env python3
3"""Module containing the RegressionPredict class and the command line interface."""
4import argparse
5import pandas as pd
6import joblib
7from biobb_common.generic.biobb_object import BiobbObject
8from sklearn.preprocessing import StandardScaler, PolynomialFeatures
9from sklearn import linear_model
10from sklearn import ensemble
11from biobb_common.configuration import settings
12from biobb_common.tools import file_utils as fu
13from biobb_common.tools.file_utils import launchlogger
14from biobb_ml.regression.common import check_input_path, check_output_path, getHeader, get_list_of_predictors, get_keys_of_predictors
17class RegressionPredict(BiobbObject):
18 """
19 | biobb_ml RegressionPredict
20 | Makes predictions from an input dataset and a given regression model.
21 | Makes predictions from an input dataset (provided either as a file or as a dictionary property) and a given regression model trained with `LinearRegression <https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html>`_, `RandomForestRegressor <https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html>`_ methods.
23 Args:
24 input_model_path (str): Path to the input model. File type: input. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/data/regression/model_regression_predict.pkl>`_. Accepted formats: pkl (edam:format_3653).
25 input_dataset_path (str) (Optional): Path to the dataset to predict. File type: input. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/data/regression/input_regression_predict.csv>`_. Accepted formats: csv (edam:format_3752).
26 output_results_path (str): Path to the output results file. File type: output. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/reference/regression/ref_output_regression_predict.csv>`_. Accepted formats: csv (edam:format_3752).
27 properties (dic - Python dictionary object containing the tool parameters, not input/output files):
28 * **predictions** (*list*) - (None) List of dictionaries with all values you want to predict targets. It will be taken into account only in case **input_dataset_path** is not provided. Format: [{ 'var1': 1.0, 'var2': 2.0 }, { 'var1': 4.0, 'var2': 2.7 }] for datasets with headers and [[ 1.0, 2.0 ], [ 4.0, 2.7 ]] for datasets without headers.
29 * **remove_tmp** (*bool*) - (True) [WF property] Remove temporal files.
30 * **restart** (*bool*) - (False) [WF property] Do not execute if output files exist.
32 Examples:
33 This is a use example of how to use the building block from Python::
35 from biobb_ml.regression.regression_predict import regression_predict
36 prop = {
37 'predictions': [
38 {
39 'var1': 1.0,
40 'var2': 2.0
41 },
42 {
43 'var1': 4.0,
44 'var2': 2.7
45 }
46 ]
47 }
48 regression_predict(input_model_path='/path/to/myModel.pkl',
49 output_results_path='/path/to/newPredictedResults.csv',
50 input_dataset_path='/path/to/myDataset.csv',
51 properties=prop)
53 Info:
54 * wrapped_software:
55 * name: scikit-learn
56 * version: >=0.24.2
57 * license: BSD 3-Clause
58 * ontology:
59 * name: EDAM
60 * schema: http://edamontology.org/EDAM.owl
62 """
64 def __init__(self, input_model_path, output_results_path,
65 input_dataset_path=None, properties=None, **kwargs) -> None:
66 properties = properties or {}
68 # Call parent class constructor
69 super().__init__(properties)
70 self.locals_var_dict = locals().copy()
72 # Input/Output files
73 self.io_dict = {
74 "in": {"input_model_path": input_model_path, "input_dataset_path": input_dataset_path},
75 "out": {"output_results_path": output_results_path}
76 }
78 # Properties specific for BB
79 self.predictions = properties.get('predictions', [])
80 self.properties = properties
82 # Check the properties
83 self.check_properties(properties)
84 self.check_arguments()
86 def check_data_params(self, out_log, err_log):
87 """ Checks all the input/output paths and parameters """
88 self.io_dict["in"]["input_model_path"] = check_input_path(self.io_dict["in"]["input_model_path"], "input_model_path", out_log, self.__class__.__name__)
89 self.io_dict["out"]["output_results_path"] = check_output_path(self.io_dict["out"]["output_results_path"], "output_results_path", False, out_log, self.__class__.__name__)
90 if self.io_dict["in"]["input_dataset_path"]:
91 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__)
93 @launchlogger
94 def launch(self) -> int:
95 """Execute the :class:`RegressionPredict <regression.regression_predict.RegressionPredict>` regression.regression_predict.RegressionPredict object."""
97 # check input/output paths and parameters
98 self.check_data_params(self.out_log, self.err_log)
100 # Setup Biobb
101 if self.check_restart():
102 return 0
103 self.stage_files()
105 fu.log('Getting model from %s' % self.io_dict["in"]["input_model_path"], self.out_log, self.global_log)
107 with open(self.io_dict["in"]["input_model_path"], "rb") as f:
108 while True:
109 try:
110 m = joblib.load(f)
111 if (isinstance(m, linear_model.LinearRegression) or isinstance(m, ensemble.RandomForestRegressor)):
112 new_model = m
113 if isinstance(m, StandardScaler):
114 scaler = m
115 if isinstance(m, PolynomialFeatures):
116 poly_features = m
117 if isinstance(m, dict):
118 variables = m
119 except EOFError:
120 break
122 if self.io_dict["in"]["input_dataset_path"]:
123 # load dataset from input_dataset_path file
124 fu.log('Getting dataset from %s' % self.io_dict["in"]["input_dataset_path"], self.out_log, self.global_log)
125 if 'columns' in variables['independent_vars']:
126 labels = getHeader(self.io_dict["in"]["input_dataset_path"])
127 skiprows = 1
128 else:
129 labels = None
130 skiprows = None
131 new_data_table = pd.read_csv(self.io_dict["in"]["input_dataset_path"], header=None, sep="\\s+|;|:|,|\t", engine="python", skiprows=skiprows, names=labels)
132 else:
133 # load dataset from properties
134 if 'columns' in variables['independent_vars']:
135 # sorting self.properties in the correct order given by variables['independent_vars']['columns']
136 index_map = {v: i for i, v in enumerate(variables['independent_vars']['columns'])}
137 predictions = []
138 for i, pred in enumerate(self.predictions):
139 sorted_pred = sorted(pred.items(), key=lambda pair: index_map[pair[0]])
140 predictions.append(dict(sorted_pred))
141 new_data_table = pd.DataFrame(data=get_list_of_predictors(predictions), columns=get_keys_of_predictors(predictions))
142 else:
143 predictions = self.predictions
144 new_data_table = pd.DataFrame(data=predictions)
146 if variables['scale']:
147 fu.log('Scaling dataset', self.out_log, self.global_log)
148 new_data = scaler.transform(new_data_table)
149 else:
150 new_data = new_data_table
152 if 'poly_features' in locals():
153 new_data = poly_features.transform(new_data)
154 p = new_model.predict(new_data)
156 if self.io_dict["in"]["input_dataset_path"] or 'columns' in variables['independent_vars']:
157 new_data_table[variables['target']['column']] = p
158 else:
159 new_data_table[len(new_data_table.columns)] = p
161 fu.log('Predicting results\n\nPREDICTION RESULTS\n\n%s\n' % new_data_table, self.out_log, self.global_log)
162 fu.log('Saving results to %s' % self.io_dict["out"]["output_results_path"], self.out_log, self.global_log)
163 new_data_table.to_csv(self.io_dict["out"]["output_results_path"], index=False, header=True, float_format='%.3f')
165 # Copy files to host
166 self.copy_to_host()
168 self.tmp_files.extend([
169 self.stage_io_dict.get("unique_dir")
170 ])
171 self.remove_tmp_files()
173 self.check_arguments(output_files_created=True, raise_exception=False)
175 return 0
178def regression_predict(input_model_path: str, output_results_path: str, input_dataset_path: str = None, properties: dict = None, **kwargs) -> int:
179 """Execute the :class:`RegressionPredict <regression.regression_predict.RegressionPredict>` class and
180 execute the :meth:`launch() <regression.regression_predict.RegressionPredict.launch>` method."""
182 return RegressionPredict(input_model_path=input_model_path,
183 output_results_path=output_results_path,
184 input_dataset_path=input_dataset_path,
185 properties=properties, **kwargs).launch()
188def main():
189 """Command line execution of this building block. Please check the command line documentation."""
190 parser = argparse.ArgumentParser(description="Makes predictions from an input dataset and a given regression model.", formatter_class=lambda prog: argparse.RawTextHelpFormatter(prog, width=99999))
191 parser.add_argument('--config', required=False, help='Configuration file')
193 # Specific args of each building block
194 required_args = parser.add_argument_group('required arguments')
195 required_args.add_argument('--input_model_path', required=True, help='Path to the input model. Accepted formats: pkl.')
196 required_args.add_argument('--output_results_path', required=True, help='Path to the output results file. Accepted formats: csv.')
197 parser.add_argument('--input_dataset_path', required=False, help='Path to the dataset to predict. Accepted formats: csv.')
199 args = parser.parse_args()
200 args.config = args.config or "{}"
201 properties = settings.ConfReader(config=args.config).get_prop_dic()
203 # Specific call of each building block
204 regression_predict(input_model_path=args.input_model_path,
205 output_results_path=args.output_results_path,
206 input_dataset_path=args.input_dataset_path,
207 properties=properties)
210if __name__ == '__main__':
211 main()