Coverage for biobb_ml/regression/random_forest_regressor.py: 85%
144 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 RandomForestRegressor class and the command line interface."""
4import argparse
5import joblib
6import numpy as np
7import pandas as pd
8from biobb_common.generic.biobb_object import BiobbObject
9from sklearn.preprocessing import StandardScaler
10from sklearn.model_selection import train_test_split
11from sklearn.metrics import mean_squared_error, r2_score
12from sklearn import ensemble
13from biobb_common.configuration import settings
14from biobb_common.tools import file_utils as fu
15from biobb_common.tools.file_utils import launchlogger
16from biobb_ml.regression.common import check_input_path, check_output_path, getHeader, getIndependentVars, getIndependentVarsList, getTarget, getTargetValue, getWeight, plotResults
19class RandomForestRegressor(BiobbObject):
20 """
21 | biobb_ml RandomForestRegressor
22 | Wrapper of the scikit-learn RandomForestRegressor method.
23 | Trains and tests a given dataset and saves the model and scaler. Visit the `RandomForestRegressor documentation page <https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html>`_.
25 Args:
26 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_random_forest_regressor.csv>`_. Accepted formats: csv (edam:format_3752).
27 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_random_forest_regressor.pkl>`_. Accepted formats: pkl (edam:format_3653).
28 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_random_forest_regressor.csv>`_. Accepted formats: csv (edam:format_3752).
29 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_random_forest_regressor.png>`_. Accepted formats: png (edam:format_3603).
30 properties (dic - Python dictionary object containing the tool parameters, not input/output files):
31 * **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.
32 * **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.
33 * **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.
34 * **n_estimators** (*int*) - (10) The number of trees in the forest.
35 * **max_depth** (*int*) - (None) The maximum depth of the tree.
36 * **random_state_method** (*int*) - (5) [1~1000|1] Controls the randomness of the estimator.
37 * **random_state_train_test** (*int*) - (5) [1~1000|1] Controls the shuffling applied to the data before applying the split.
38 * **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.
39 * **scale** (*bool*) - (False) Whether or not to scale the input dataset.
40 * **remove_tmp** (*bool*) - (True) [WF property] Remove temporal files.
41 * **restart** (*bool*) - (False) [WF property] Do not execute if output files exist.
43 Examples:
44 This is a use example of how to use the building block from Python::
46 from biobb_ml.regression.random_forest_regressor import random_forest_regressor
47 prop = {
48 'independent_vars': {
49 'columns': [ 'column1', 'column2', 'column3' ]
50 },
51 'target': {
52 'column': 'target'
53 },
54 'n_estimators': 10,
55 'max_depth': 5,
56 'test_size': 0.2
57 }
58 random_forest_regressor(input_dataset_path='/path/to/myDataset.csv',
59 output_model_path='/path/to/newModel.pkl',
60 output_test_table_path='/path/to/newTable.csv',
61 output_plot_path='/path/to/newPlot.png',
62 properties=prop)
64 Info:
65 * wrapped_software:
66 * name: scikit-learn RandomForestRegressor
67 * version: >0.24.2
68 * license: BSD 3-Clause
69 * ontology:
70 * name: EDAM
71 * schema: http://edamontology.org/EDAM.owl
73 """
75 def __init__(self, input_dataset_path, output_model_path,
76 output_test_table_path=None, output_plot_path=None, properties=None, **kwargs) -> None:
77 properties = properties or {}
79 # Call parent class constructor
80 super().__init__(properties)
81 self.locals_var_dict = locals().copy()
83 # Input/Output files
84 self.io_dict = {
85 "in": {"input_dataset_path": input_dataset_path},
86 "out": {"output_model_path": output_model_path, "output_test_table_path": output_test_table_path, "output_plot_path": output_plot_path}
87 }
89 # Properties specific for BB
90 self.independent_vars = properties.get('independent_vars', {})
91 self.target = properties.get('target', {})
92 self.weight = properties.get('weight', {})
93 self.n_estimators = properties.get('n_estimators', 10)
94 self.max_depth = properties.get('max_depth', None)
95 self.random_state_method = properties.get('random_state_method', 5)
96 self.random_state_train_test = properties.get('random_state_train_test', 5)
97 self.test_size = properties.get('test_size', 0.2)
98 self.scale = properties.get('scale', False)
99 self.properties = properties
101 # Check the properties
102 self.check_properties(properties)
103 self.check_arguments()
105 def check_data_params(self, out_log, err_log):
106 """ Checks all the input/output paths and parameters """
107 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__)
108 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__)
109 if self.io_dict["out"]["output_test_table_path"]:
110 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__)
111 if self.io_dict["out"]["output_plot_path"]:
112 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__)
114 @launchlogger
115 def launch(self) -> int:
116 """Execute the :class:`RandomForestRegressor <regression.random_forest_regressor.RandomForestRegressor>` regression.random_forest_regressor.RandomForestRegressor object."""
118 # check input/output paths and parameters
119 self.check_data_params(self.out_log, self.err_log)
121 # Setup Biobb
122 if self.check_restart():
123 return 0
124 self.stage_files()
126 # load dataset
127 fu.log('Getting dataset from %s' % self.io_dict["in"]["input_dataset_path"], self.out_log, self.global_log)
128 if 'columns' in self.independent_vars:
129 labels = getHeader(self.io_dict["in"]["input_dataset_path"])
130 skiprows = 1
131 else:
132 labels = None
133 skiprows = None
134 data = pd.read_csv(self.io_dict["in"]["input_dataset_path"], header=None, sep="\\s+|;|:|,|\t", engine="python", skiprows=skiprows, names=labels)
136 # declare inputs, targets and weights
137 # the inputs are all the independent variables
138 X = getIndependentVars(self.independent_vars, data, self.out_log, self.__class__.__name__)
139 fu.log('Independent variables: [%s]' % (getIndependentVarsList(self.independent_vars)), self.out_log, self.global_log)
140 # target
141 y = getTarget(self.target, data, self.out_log, self.__class__.__name__)
142 fu.log('Target: %s' % (getTargetValue(self.target)), self.out_log, self.global_log)
143 # weights
144 if self.weight:
145 w = getWeight(self.weight, data, self.out_log, self.__class__.__name__)
146 fu.log('Weight column provided', self.out_log, self.global_log)
148 # train / test split
149 fu.log('Creating train and test sets', self.out_log, self.global_log)
150 arrays_sets = (X, y)
151 # if user provide weights
152 if self.weight:
153 arrays_sets = arrays_sets + (w,)
154 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)
155 else:
156 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)
158 # scale dataset
159 if self.scale:
160 fu.log('Scaling dataset', self.out_log, self.global_log)
161 scaler = StandardScaler()
162 X_train = scaler.fit_transform(X_train)
164 # regression
165 fu.log('Training dataset applying random forest regressor', self.out_log, self.global_log)
166 model = ensemble.RandomForestRegressor(max_depth=self.max_depth, n_estimators=self.n_estimators, random_state=self.random_state_method)
167 arrays_fit = (X_train, y_train)
168 # if user provide weights
169 if self.weight:
170 arrays_fit = arrays_fit + (w_train,)
172 model.fit(*arrays_fit)
174 # scores and coefficients train
175 y_hat_train = model.predict(X_train)
176 rmse = (np.sqrt(mean_squared_error(y_train, y_hat_train)))
177 rss = ((y_train - y_hat_train) ** 2).sum()
178 score_train_inputs = (y_train, y_hat_train)
179 if self.weight:
180 score_train_inputs = score_train_inputs + (w_train,)
181 score = r2_score(*score_train_inputs)
183 # r-squared
184 r2_table = pd.DataFrame()
185 r2_table["feature"] = ['R2', 'RMSE', 'RSS']
186 r2_table['coefficient'] = [score, rmse, rss]
188 fu.log('Calculating scores and coefficients for TRAINING dataset\n\nSCORES\n\n%s\n' % r2_table, self.out_log, self.global_log)
190 # testing
191 # predict data from x_test
192 if self.scale:
193 X_test = scaler.transform(X_test)
194 y_hat_test = model.predict(X_test)
195 test_table = pd.DataFrame(y_hat_test, columns=['prediction'])
196 # reset y_test (problem with old indexes column)
197 y_test = y_test.reset_index(drop=True)
198 # add real values to predicted ones in test_table table
199 test_table['target'] = y_test
200 # calculate difference between target and prediction (absolute and %)
201 test_table['residual'] = test_table['target'] - test_table['prediction']
202 test_table['difference %'] = np.absolute(test_table['residual']/test_table['target']*100)
203 # sort by difference in %
204 test_table = test_table.sort_values(by=['difference %'])
205 test_table = test_table.reset_index(drop=True)
206 fu.log('Testing\n\nTEST DATA\n\n%s\n' % test_table, self.out_log, self.global_log)
208 # scores and coefficients test
209 score_test_inputs = (y_test, y_hat_test)
210 if self.weight:
211 score_test_inputs = score_test_inputs + (w_test,)
212 r2_test = r2_score(*score_test_inputs)
213 rmse_test = np.sqrt(mean_squared_error(y_test, y_hat_test))
214 rss_test = ((y_test - y_hat_test) ** 2).sum()
216 # r-squared
217 r2_table_test = pd.DataFrame()
218 r2_table_test["feature"] = ['R2', 'RMSE', 'RSS']
219 r2_table_test['coefficient'] = [r2_test, rmse_test, rss_test]
221 fu.log('Calculating scores and coefficients for TESTING dataset\n\nSCORES\n\n%s\n' % r2_table_test, self.out_log, self.global_log)
223 if (self.io_dict["out"]["output_test_table_path"]):
224 fu.log('Saving testing data to %s' % self.io_dict["out"]["output_test_table_path"], self.out_log, self.global_log)
225 test_table.to_csv(self.io_dict["out"]["output_test_table_path"], index=False, header=True)
227 # create test plot
228 if (self.io_dict["out"]["output_plot_path"]):
229 fu.log('Saving residual plot to %s' % self.io_dict["out"]["output_plot_path"], self.out_log, self.global_log)
230 y_hat_test = y_hat_test.flatten()
231 y_hat_train = y_hat_train.flatten()
232 plot = plotResults(y_train, y_hat_train, y_test, y_hat_test)
233 plot.savefig(self.io_dict["out"]["output_plot_path"], dpi=150)
235 # save model, scaler and parameters
236 variables = {
237 'target': self.target,
238 'independent_vars': self.independent_vars,
239 'scale': self.scale
240 }
241 fu.log('Saving model to %s' % self.io_dict["out"]["output_model_path"], self.out_log, self.global_log)
242 with open(self.io_dict["out"]["output_model_path"], "wb") as f:
243 joblib.dump(model, f)
244 if self.scale:
245 joblib.dump(scaler, f)
246 joblib.dump(variables, f)
248 # Copy files to host
249 self.copy_to_host()
251 self.tmp_files.extend([
252 self.stage_io_dict.get("unique_dir")
253 ])
254 self.remove_tmp_files()
256 self.check_arguments(output_files_created=True, raise_exception=False)
258 return 0
261def random_forest_regressor(input_dataset_path: str, output_model_path: str, output_test_table_path: str = None, output_plot_path: str = None, properties: dict = None, **kwargs) -> int:
262 """Execute the :class:`RandomForestRegressor <regression.random_forest_regressor.RandomForestRegressor>` class and
263 execute the :meth:`launch() <regression.random_forest_regressor.RandomForestRegressor.launch>` method."""
265 return RandomForestRegressor(input_dataset_path=input_dataset_path,
266 output_model_path=output_model_path,
267 output_test_table_path=output_test_table_path,
268 output_plot_path=output_plot_path,
269 properties=properties, **kwargs).launch()
272def main():
273 """Command line execution of this building block. Please check the command line documentation."""
274 parser = argparse.ArgumentParser(description="Wrapper of the scikit-learn RandomForestRegressor method.", formatter_class=lambda prog: argparse.RawTextHelpFormatter(prog, width=99999))
275 parser.add_argument('--config', required=False, help='Configuration file')
277 # Specific args of each building block
278 required_args = parser.add_argument_group('required arguments')
279 required_args.add_argument('--input_dataset_path', required=True, help='Path to the input dataset. Accepted formats: csv.')
280 required_args.add_argument('--output_model_path', required=True, help='Path to the output model file. Accepted formats: pkl.')
281 parser.add_argument('--output_test_table_path', required=False, help='Path to the test table file. Accepted formats: csv.')
282 parser.add_argument('--output_plot_path', required=False, help='Residual plot checks the error between actual values and predicted values. Accepted formats: png.')
284 args = parser.parse_args()
285 args.config = args.config or "{}"
286 properties = settings.ConfReader(config=args.config).get_prop_dic()
288 # Specific call of each building block
289 random_forest_regressor(input_dataset_path=args.input_dataset_path,
290 output_model_path=args.output_model_path,
291 output_test_table_path=args.output_test_table_path,
292 output_plot_path=args.output_plot_path,
293 properties=properties)
296if __name__ == '__main__':
297 main()