Coverage for biobb_ml/regression/polynomial_regression.py: 87%
146 statements
« prev ^ index » next coverage.py v7.5.1, created at 2024-05-07 09:39 +0000
« 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 PolynomialRegression 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, PolynomialFeatures
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 PolynomialRegression(BiobbObject):
21 """
22 | biobb_ml PolynomialRegression
23 | Wrapper of the scikit-learn LinearRegression method with PolynomialFeatures.
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_polynomial_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_polynomial_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_polynomial_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_polynomial_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 * **degree** (*int*) - (2) [1~100|1] Polynomial degree.
37 * **test_size** (*float*) - (0.2) Represents the proportion of the dataset to include in the test split. It should be between 0.0 and 1.0.
38 * **scale** (*bool*) - (False) Whether or not to scale the input dataset.
39 * **remove_tmp** (*bool*) - (True) [WF property] Remove temporal files.
40 * **restart** (*bool*) - (False) [WF property] Do not execute if output files exist.
42 Examples:
43 This is a use example of how to use the building block from Python::
45 from biobb_ml.regression.polynomial_regression import polynomial_regression
46 prop = {
47 'independent_vars': {
48 'columns': [ 'column1', 'column2', 'column3' ]
49 },
50 'target': {
51 'column': 'target'
52 },
53 'degree': 2,
54 'test_size': 0.2
55 }
56 polynomial_regression(input_dataset_path='/path/to/myDataset.csv',
57 output_model_path='/path/to/newModel.pkl',
58 output_test_table_path='/path/to/newTable.csv',
59 output_plot_path='/path/to/newPlot.png',
60 properties=prop)
62 Info:
63 * wrapped_software:
64 * name: scikit-learn LinearRegression
65 * version: >=0.24.2
66 * license: BSD 3-Clause
67 * ontology:
68 * name: EDAM
69 * schema: http://edamontology.org/EDAM.owl
71 """
73 def __init__(self, input_dataset_path, output_model_path,
74 output_test_table_path=None, output_plot_path=None, properties=None, **kwargs) -> None:
75 properties = properties or {}
77 # Call parent class constructor
78 super().__init__(properties)
79 self.locals_var_dict = locals().copy()
81 # Input/Output files
82 self.io_dict = {
83 "in": {"input_dataset_path": input_dataset_path},
84 "out": {"output_model_path": output_model_path, "output_test_table_path": output_test_table_path, "output_plot_path": output_plot_path}
85 }
87 # Properties specific for BB
88 self.independent_vars = properties.get('independent_vars', {})
89 self.target = properties.get('target', {})
90 self.weight = properties.get('weight', {})
91 self.random_state_train_test = properties.get('random_state_train_test', 5)
92 self.degree = properties.get('degree', 2)
93 self.test_size = properties.get('test_size', 0.2)
94 self.scale = properties.get('scale', False)
95 self.properties = properties
97 # Check the properties
98 self.check_properties(properties)
99 self.check_arguments()
101 def check_data_params(self, out_log, err_log):
102 """ Checks all the input/output paths and parameters """
103 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__)
104 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__)
105 if self.io_dict["out"]["output_test_table_path"]:
106 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__)
107 if self.io_dict["out"]["output_plot_path"]:
108 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__)
110 @launchlogger
111 def launch(self) -> int:
112 """Execute the :class:`PolynomialRegression <regression.polynomial_regression.PolynomialRegression>` regression.polynomial_regression.PolynomialRegression object."""
114 # check input/output paths and parameters
115 self.check_data_params(self.out_log, self.err_log)
117 # Setup Biobb
118 if self.check_restart():
119 return 0
120 self.stage_files()
122 # load dataset
123 fu.log('Getting dataset from %s' % self.io_dict["in"]["input_dataset_path"], self.out_log, self.global_log)
124 if 'columns' in self.independent_vars:
125 labels = getHeader(self.io_dict["in"]["input_dataset_path"])
126 skiprows = 1
127 else:
128 labels = None
129 skiprows = None
130 data = pd.read_csv(self.io_dict["in"]["input_dataset_path"], header=None, sep="\\s+|;|:|,|\t", engine="python", skiprows=skiprows, names=labels)
132 # declare inputs, targets and weights
133 # the inputs are all the independent variables
134 X = getIndependentVars(self.independent_vars, data, self.out_log, self.__class__.__name__)
135 fu.log('Independent variables: [%s]' % (getIndependentVarsList(self.independent_vars)), self.out_log, self.global_log)
136 # target
137 y = getTarget(self.target, data, self.out_log, self.__class__.__name__)
138 fu.log('Target: %s' % (getTargetValue(self.target)), self.out_log, self.global_log)
139 # weights
140 if self.weight:
141 w = getWeight(self.weight, data, self.out_log, self.__class__.__name__)
142 fu.log('Weight column provided', self.out_log, self.global_log)
144 # train / test split
145 fu.log('Creating train and test sets', self.out_log, self.global_log)
146 arrays_sets = (X, y)
147 # if user provide weights
148 if self.weight:
149 arrays_sets = arrays_sets + (w,)
150 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)
151 else:
152 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)
154 # scale dataset
155 if self.scale:
156 fu.log('Scaling dataset', self.out_log, self.global_log)
157 scaler = StandardScaler()
158 X_train = scaler.fit_transform(X_train)
160 # regression
161 fu.log('Training dataset applying polynomial regression', self.out_log, self.global_log)
162 poly_features = PolynomialFeatures(degree=self.degree)
163 X_train_poly = poly_features.fit_transform(X_train)
164 model = linear_model.LinearRegression()
165 model.fit(X_train_poly, y_train)
167 # scores and coefficients train
168 y_hat_train = model.predict(X_train_poly)
169 rmse = (np.sqrt(mean_squared_error(y_train, y_hat_train)))
170 rss = ((y_train - y_hat_train) ** 2).sum()
171 score = r2_score(y_hat_train, y_train)
172 # bias = model.intercept_
173 coef = model.coef_
174 coef = ['%.3f' % item for item in coef]
175 adj_r2 = adjusted_r2(X_train_poly, y_train, score)
176 p_values = f_regression(X_train_poly, y_train)[1]
177 p_values = ['%.3f' % item for item in p_values]
179 # r-squared
180 r2_table = pd.DataFrame()
181 r2_table["feature"] = ['R2', 'Adj. R2', 'RMSE', 'RSS']
182 r2_table['coefficient'] = [score, adj_r2, rmse, rss]
184 fu.log('Calculating scores and coefficients for training dataset\n\nR2, ADJUSTED R2 & RMSE\n\n%s\n' % r2_table, self.out_log, self.global_log)
186 if self.scale:
187 X_test = scaler.transform(X_test)
188 X_test_poly = poly_features.fit_transform(X_test)
189 y_hat_test = model.predict(X_test_poly)
190 test_table = pd.DataFrame(y_hat_test, columns=['prediction'])
191 # reset y_test (problem with old indexes column)
192 y_test = y_test.reset_index(drop=True)
193 # add real values to predicted ones in test_table table
194 test_table['target'] = y_test
195 # calculate difference between target and prediction (absolute and %)
196 test_table['residual'] = test_table['target'] - test_table['prediction']
197 test_table['difference %'] = np.absolute(test_table['residual']/test_table['target']*100)
198 pd.set_option('display.float_format', lambda x: '%.2f' % x)
199 # sort by difference in %
200 test_table = test_table.sort_values(by=['difference %'])
201 test_table = test_table.reset_index(drop=True)
202 fu.log('Testing\n\nTEST DATA\n\n%s\n' % test_table, self.out_log, self.global_log)
204 # scores and coefficients test
205 r2_test = r2_score(y_hat_test, y_test)
206 adj_r2_test = adjusted_r2(X_test_poly, y_test, r2_test)
207 rmse_test = np.sqrt(mean_squared_error(y_test, y_hat_test))
208 rss_test = ((y_test - y_hat_test) ** 2).sum()
210 # r-squared
211 pd.set_option('display.float_format', lambda x: '%.6f' % x)
212 r2_table_test = pd.DataFrame()
213 r2_table_test["feature"] = ['R2', 'Adj. R2', 'RMSE', 'RSS']
214 r2_table_test['coefficient'] = [r2_test, adj_r2_test, rmse_test, rss_test]
216 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)
218 if (self.io_dict["out"]["output_test_table_path"]):
219 fu.log('Saving testing data to %s' % self.io_dict["out"]["output_test_table_path"], self.out_log, self.global_log)
220 test_table.to_csv(self.io_dict["out"]["output_test_table_path"], index=False, header=True)
222 # create test plot
223 if (self.io_dict["out"]["output_plot_path"]):
224 fu.log('Saving residual plot to %s' % self.io_dict["out"]["output_plot_path"], self.out_log, self.global_log)
225 y_hat_test = y_hat_test.flatten()
226 y_hat_train = y_hat_train.flatten()
227 plot = plotResults(y_train, y_hat_train, y_test, y_hat_test)
228 plot.savefig(self.io_dict["out"]["output_plot_path"], dpi=150)
230 # save model, scaler and parameters
231 variables = {
232 'target': self.target,
233 'independent_vars': self.independent_vars,
234 'scale': self.scale
235 }
236 fu.log('Saving model to %s' % self.io_dict["out"]["output_model_path"], self.out_log, self.global_log)
237 with open(self.io_dict["out"]["output_model_path"], "wb") as f:
238 joblib.dump(model, f)
239 if self.scale:
240 joblib.dump(scaler, f)
241 joblib.dump(poly_features, f)
242 joblib.dump(variables, f)
244 # Copy files to host
245 self.copy_to_host()
247 self.tmp_files.extend([
248 self.stage_io_dict.get("unique_dir")
249 ])
250 self.remove_tmp_files()
252 self.check_arguments(output_files_created=True, raise_exception=False)
254 return 0
257def polynomial_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:
258 """Execute the :class:`PolynomialRegression <regression.polynomial_regression.PolynomialRegression>` class and
259 execute the :meth:`launch() <regression.polynomial_regression.PolynomialRegression.launch>` method."""
261 return PolynomialRegression(input_dataset_path=input_dataset_path,
262 output_model_path=output_model_path,
263 output_test_table_path=output_test_table_path,
264 output_plot_path=output_plot_path,
265 properties=properties, **kwargs).launch()
268def main():
269 """Command line execution of this building block. Please check the command line documentation."""
270 parser = argparse.ArgumentParser(description="Wrapper of the scikit-learn LinearRegression method with PolynomialFeatures.", formatter_class=lambda prog: argparse.RawTextHelpFormatter(prog, width=99999))
271 parser.add_argument('--config', required=False, help='Configuration file')
273 # Specific args of each building block
274 required_args = parser.add_argument_group('required arguments')
275 required_args.add_argument('--input_dataset_path', required=True, help='Path to the input dataset. Accepted formats: csv.')
276 required_args.add_argument('--output_model_path', required=True, help='Path to the output model file. Accepted formats: pkl.')
277 parser.add_argument('--output_test_table_path', required=False, help='Path to the test table file. Accepted formats: csv.')
278 parser.add_argument('--output_plot_path', required=False, help='Residual plot checks the error between actual values and predicted values. Accepted formats: png.')
280 args = parser.parse_args()
281 args.config = args.config or "{}"
282 properties = settings.ConfReader(config=args.config).get_prop_dic()
284 # Specific call of each building block
285 polynomial_regression(input_dataset_path=args.input_dataset_path,
286 output_model_path=args.output_model_path,
287 output_test_table_path=args.output_test_table_path,
288 output_plot_path=args.output_plot_path,
289 properties=properties)
292if __name__ == '__main__':
293 main()