Coverage for biobb_ml/regression/linear_regression.py: 87%
154 statements
« prev ^ index » next coverage.py v7.6.1, created at 2024-10-03 14:57 +0000
1#!/usr/bin/env python3
3"""Module containing the LinearRegression class and the command line interface."""
4import argparse
5import numpy as np
6import pandas as pd
7import joblib
8from biobb_common.generic.biobb_object import BiobbObject
9from sklearn.preprocessing import StandardScaler
10from sklearn.model_selection import train_test_split
11from sklearn.feature_selection import f_regression
12from sklearn.metrics import mean_squared_error, r2_score
13from sklearn import linear_model
14from biobb_common.configuration import settings
15from biobb_common.tools import file_utils as fu
16from biobb_common.tools.file_utils import launchlogger
17from biobb_ml.regression.common import check_input_path, check_output_path, getHeader, getIndependentVars, getIndependentVarsList, getTarget, getTargetValue, getWeight, adjusted_r2, plotResults
20class LinearRegression(BiobbObject):
21 """
22 | biobb_ml LinearRegression
23 | Wrapper of the scikit-learn LinearRegression method.
24 | Trains and tests a given dataset and saves the model and scaler. Visit the `LinearRegression documentation page <https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html>`_ in the sklearn official website for further information.
26 Args:
27 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/regression/dataset_linear_regression.csv>`_. Accepted formats: csv (edam:format_3752).
28 output_model_path (str): Path to the output model file. File type: output. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/reference/regression/ref_output_model_linear_regression.pkl>`_. Accepted formats: pkl (edam:format_3653).
29 output_test_table_path (str) (Optional): Path to the test table file. File type: output. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/reference/regression/ref_output_test_linear_regression.csv>`_. Accepted formats: csv (edam:format_3752).
30 output_plot_path (str) (Optional): Residual plot checks the error between actual values and predicted values. File type: output. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/reference/regression/ref_output_plot_linear_regression.png>`_. Accepted formats: png (edam:format_3603).
31 properties (dic - Python dictionary object containing the tool parameters, not input/output files):
32 * **independent_vars** (*dict*) - ({}) Independent variables you want to train from your dataset. 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.
33 * **target** (*dict*) - ({}) Dependent variable you want to predict from your dataset. You can specify either a column name or a column index. Formats: { "column": "column3" } or { "index": 21 }. In case of mulitple formats, the first one will be picked.
34 * **weight** (*dict*) - ({}) Weight variable from your dataset. You can specify either a column name or a column index. Formats: { "column": "column3" } or { "index": 21 }. In case of mulitple formats, the first one will be picked.
35 * **random_state_train_test** (*int*) - (5) [1~1000|1] Controls the shuffling applied to the data before applying the split.
36 * **test_size** (*float*) - (0.2) [0~1|0.05] Represents the proportion of the dataset to include in the test split. It should be between 0.0 and 1.0.
37 * **scale** (*bool*) - (False) Whether or not to scale the input dataset.
38 * **remove_tmp** (*bool*) - (True) [WF property] Remove temporal files.
39 * **restart** (*bool*) - (False) [WF property] Do not execute if output files exist.
40 * **sandbox_path** (*str*) - ("./") [WF property] Parent path to the sandbox directory.
42 Examples:
43 This is a use example of how to use the building block from Python::
45 from biobb_ml.regression.linear_regression import linear_regression
46 prop = {
47 'independent_vars': {
48 'columns': [ 'column1', 'column2', 'column3' ]
49 },
50 'target': {
51 'column': 'target'
52 },
53 'test_size': 0.2
54 }
55 linear_regression(input_dataset_path='/path/to/myDataset.csv',
56 output_model_path='/path/to/newModel.pkl',
57 output_test_table_path='/path/to/newTable.csv',
58 output_plot_path='/path/to/newPlot.png',
59 properties=prop)
61 Info:
62 * wrapped_software:
63 * name: scikit-learn LinearRegression
64 * version: >=0.24.2
65 * license: BSD 3-Clause
66 * ontology:
67 * name: EDAM
68 * schema: http://edamontology.org/EDAM.owl
70 """
72 def __init__(self, input_dataset_path, output_model_path,
73 output_test_table_path=None, output_plot_path=None, properties=None, **kwargs) -> None:
74 properties = properties or {}
76 # Call parent class constructor
77 super().__init__(properties)
78 self.locals_var_dict = locals().copy()
80 # Input/Output files
81 self.io_dict = {
82 "in": {"input_dataset_path": input_dataset_path},
83 "out": {"output_model_path": output_model_path, "output_test_table_path": output_test_table_path, "output_plot_path": output_plot_path}
84 }
86 # Properties specific for BB
87 self.independent_vars = properties.get('independent_vars', {})
88 self.target = properties.get('target', {})
89 self.weight = properties.get('weight', {})
90 self.random_state_train_test = properties.get('random_state_train_test', 5)
91 self.test_size = properties.get('test_size', 0.2)
92 self.scale = properties.get('scale', False)
93 self.properties = properties
95 # Check the properties
96 self.check_properties(properties)
97 self.check_arguments()
99 def check_data_params(self, out_log, err_log):
100 """ Checks all the input/output paths and parameters """
101 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__)
102 self.io_dict["out"]["output_model_path"] = check_output_path(self.io_dict["out"]["output_model_path"], "output_model_path", False, out_log, self.__class__.__name__)
103 if self.io_dict["out"]["output_test_table_path"]:
104 self.io_dict["out"]["output_test_table_path"] = check_output_path(self.io_dict["out"]["output_test_table_path"], "output_test_table_path", True, out_log, self.__class__.__name__)
105 if self.io_dict["out"]["output_plot_path"]:
106 self.io_dict["out"]["output_plot_path"] = check_output_path(self.io_dict["out"]["output_plot_path"], "output_plot_path", True, out_log, self.__class__.__name__)
108 @launchlogger
109 def launch(self) -> int:
110 """Execute the :class:`LinearRegression <regression.linear_regression.LinearRegression>` regression.linear_regression.LinearRegression object."""
112 # check input/output paths and parameters
113 self.check_data_params(self.out_log, self.err_log)
115 # Setup Biobb
116 if self.check_restart():
117 return 0
118 self.stage_files()
120 # load dataset
121 fu.log('Getting dataset from %s' % self.io_dict["in"]["input_dataset_path"], self.out_log, self.global_log)
122 if 'columns' in self.independent_vars:
123 labels = getHeader(self.io_dict["in"]["input_dataset_path"])
124 skiprows = 1
125 else:
126 labels = None
127 skiprows = None
128 data = pd.read_csv(self.io_dict["in"]["input_dataset_path"], header=None, sep="\\s+|;|:|,|\t", engine="python", skiprows=skiprows, names=labels)
130 # declare inputs, targets and weights
131 # the inputs are all the independent variables
132 X = getIndependentVars(self.independent_vars, data, self.out_log, self.__class__.__name__)
133 fu.log('Independent variables: [%s]' % (getIndependentVarsList(self.independent_vars)), self.out_log, self.global_log)
134 # target
135 y = getTarget(self.target, data, self.out_log, self.__class__.__name__)
136 fu.log('Target: %s' % (getTargetValue(self.target)), self.out_log, self.global_log)
137 # weights
138 if self.weight:
139 w = getWeight(self.weight, data, self.out_log, self.__class__.__name__)
140 fu.log('Weight column provided', self.out_log, self.global_log)
142 # train / test split
143 fu.log('Creating train and test sets', self.out_log, self.global_log)
144 arrays_sets = (X, y)
145 # if user provide weights
146 if self.weight:
147 arrays_sets = arrays_sets + (w,)
148 X_train, X_test, y_train, y_test, w_train, w_test = train_test_split(*arrays_sets, test_size=self.test_size, random_state=self.random_state_train_test)
149 else:
150 X_train, X_test, y_train, y_test = train_test_split(*arrays_sets, test_size=self.test_size, random_state=self.random_state_train_test)
152 # scale dataset
153 if self.scale:
154 fu.log('Scaling dataset', self.out_log, self.global_log)
155 scaler = StandardScaler()
156 X_train = scaler.fit_transform(X_train)
158 # regression
159 fu.log('Training dataset applying linear regression', self.out_log, self.global_log)
160 model = linear_model.LinearRegression()
161 arrays_fit = (X_train, y_train)
162 # if user provide weights
163 if self.weight:
164 arrays_fit = arrays_fit + (w_train,)
166 model.fit(*arrays_fit)
168 # scores and coefficients train
169 y_hat_train = model.predict(X_train)
170 rmse = (np.sqrt(mean_squared_error(y_train, y_hat_train)))
171 rss = ((y_train - y_hat_train) ** 2).sum()
172 score = r2_score(y_hat_train, y_train)
173 bias = model.intercept_
174 coef = model.coef_
175 coef = ['%.3f' % item for item in coef]
176 adj_r2 = adjusted_r2(X_train, y_train, score)
177 p_values = f_regression(X_train, y_train)[1]
178 p_values = ['%.3f' % item for item in p_values]
180 # r-squared
181 r2_table = pd.DataFrame()
182 r2_table["feature"] = ['R2', 'Adj. R2', 'RMSE', 'RSS']
183 r2_table['coefficient'] = [score, adj_r2, rmse, rss]
185 # p-values
186 cols = ['bias']
187 cols.extend(list(getIndependentVarsList(self.independent_vars).split(", ")))
188 coefs_table = pd.DataFrame(cols, columns=['feature'])
189 c = [round(bias, 3)]
190 c.extend(coef)
191 c = list(map(float, c))
192 coefs_table['coefficient'] = c
193 p = [0]
194 p.extend(p_values)
196 coefs_table['p-value'] = p
197 fu.log('Calculating scores and coefficients for training dataset\n\nR2, ADJUSTED R2 & RMSE\n\n%s\n\nCOEFFS & P-VALUES\n\n%s\n' % (r2_table, coefs_table), self.out_log, self.global_log)
199 # testing
200 # predict data from x_test
201 if self.scale:
202 X_test = scaler.transform(X_test)
203 y_hat_test = model.predict(X_test)
204 test_table = pd.DataFrame(y_hat_test, columns=['prediction'])
205 # reset y_test (problem with old indexes column)
206 y_test = y_test.reset_index(drop=True)
207 # add real values to predicted ones in test_table table
208 test_table['target'] = y_test
209 # calculate difference between target and prediction (absolute and %)
210 test_table['residual'] = test_table['target'] - test_table['prediction']
211 test_table['difference %'] = np.absolute(test_table['residual']/test_table['target']*100)
212 pd.set_option('display.float_format', lambda x: '%.2f' % x)
213 # sort by difference in %
214 test_table = test_table.sort_values(by=['difference %'])
215 test_table = test_table.reset_index(drop=True)
216 fu.log('Testing\n\nTEST DATA\n\n%s\n' % test_table, self.out_log, self.global_log)
218 # scores and coefficients test
219 r2_test = r2_score(y_hat_test, y_test)
220 adj_r2_test = adjusted_r2(X_test, y_test, r2_test)
221 rmse_test = np.sqrt(mean_squared_error(y_test, y_hat_test))
222 rss_test = ((y_test - y_hat_test) ** 2).sum()
224 # r-squared
225 pd.set_option('display.float_format', lambda x: '%.6f' % x)
226 r2_table_test = pd.DataFrame()
227 r2_table_test["feature"] = ['R2', 'Adj. R2', 'RMSE', 'RSS']
228 r2_table_test['coefficient'] = [r2_test, adj_r2_test, rmse_test, rss_test]
230 fu.log('Calculating scores and coefficients for testing dataset\n\nR2, ADJUSTED R2 & RMSE\n\n%s\n' % r2_table_test, self.out_log, self.global_log)
232 if (self.io_dict["out"]["output_test_table_path"]):
233 fu.log('Saving testing data to %s' % self.io_dict["out"]["output_test_table_path"], self.out_log, self.global_log)
234 test_table.to_csv(self.io_dict["out"]["output_test_table_path"], index=False, header=True)
236 # create test plot
237 if (self.io_dict["out"]["output_plot_path"]):
238 fu.log('Saving residual plot to %s' % self.io_dict["out"]["output_plot_path"], self.out_log, self.global_log)
239 y_hat_test = y_hat_test.flatten()
240 y_hat_train = y_hat_train.flatten()
241 plot = plotResults(y_train, y_hat_train, y_test, y_hat_test)
242 plot.savefig(self.io_dict["out"]["output_plot_path"], dpi=150)
244 # save model, scaler and parameters
245 variables = {
246 'target': self.target,
247 'independent_vars': self.independent_vars,
248 'scale': self.scale
249 }
250 fu.log('Saving model to %s' % self.io_dict["out"]["output_model_path"], self.out_log, self.global_log)
251 with open(self.io_dict["out"]["output_model_path"], "wb") as f:
252 joblib.dump(model, f)
253 if self.scale:
254 joblib.dump(scaler, f)
255 joblib.dump(variables, f)
257 # Copy files to host
258 self.copy_to_host()
260 self.tmp_files.extend([
261 self.stage_io_dict.get("unique_dir")
262 ])
263 self.remove_tmp_files()
265 self.check_arguments(output_files_created=True, raise_exception=False)
267 return 0
270def linear_regression(input_dataset_path: str, output_model_path: str, output_test_table_path: str = None, output_plot_path: str = None, properties: dict = None, **kwargs) -> int:
271 """Execute the :class:`LinearRegression <regression.linear_regression.LinearRegression>` class and
272 execute the :meth:`launch() <regression.linear_regression.LinearRegression.launch>` method."""
274 return LinearRegression(input_dataset_path=input_dataset_path,
275 output_model_path=output_model_path,
276 output_test_table_path=output_test_table_path,
277 output_plot_path=output_plot_path,
278 properties=properties, **kwargs).launch()
281def main():
282 """Command line execution of this building block. Please check the command line documentation."""
283 parser = argparse.ArgumentParser(description="Wrapper of the scikit-learn LinearRegression method.", formatter_class=lambda prog: argparse.RawTextHelpFormatter(prog, width=99999))
284 parser.add_argument('--config', required=False, help='Configuration file')
286 # Specific args of each building block
287 required_args = parser.add_argument_group('required arguments')
288 required_args.add_argument('--input_dataset_path', required=True, help='Path to the input dataset. Accepted formats: csv.')
289 required_args.add_argument('--output_model_path', required=True, help='Path to the output model file. Accepted formats: pkl.')
290 parser.add_argument('--output_test_table_path', required=False, help='Path to the test table file. Accepted formats: csv.')
291 parser.add_argument('--output_plot_path', required=False, help='Residual plot checks the error between actual values and predicted values. Accepted formats: png.')
293 args = parser.parse_args()
294 args.config = args.config or "{}"
295 properties = settings.ConfReader(config=args.config).get_prop_dic()
297 # Specific call of each building block
298 linear_regression(input_dataset_path=args.input_dataset_path,
299 output_model_path=args.output_model_path,
300 output_test_table_path=args.output_test_table_path,
301 output_plot_path=args.output_plot_path,
302 properties=properties)
305if __name__ == '__main__':
306 main()