Coverage for biobb_ml/classification/decision_tree.py: 83%
151 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 DecisionTree class and the command line interface."""
4import argparse
5import joblib
6import pandas as pd
7import numpy as np
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 confusion_matrix, classification_report, log_loss
12from sklearn.tree import DecisionTreeClassifier
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.classification.common import check_input_path, check_output_path, getHeader, getIndependentVars, getIndependentVarsList, getTarget, getTargetValue, getWeight, plotMultipleCM, plotBinaryClassifier
19class DecisionTree(BiobbObject):
20 """
21 | biobb_ml DecisionTree
22 | Wrapper of the scikit-learn DecisionTreeClassifier method.
23 | Trains and tests a given dataset and saves the model and scaler. Visit the `DecisionTreeClassifier documentation page <https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html>`_ in the sklearn official website for further information.
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/classification/dataset_decision_tree.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/classification/ref_output_model_decision_tree.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/classification/ref_output_test_decision_tree.csv>`_. Accepted formats: csv (edam:format_3752).
29 output_plot_path (str) (Optional): Path to the statistics plot. If target is binary it shows confusion matrix, distributions of the predicted probabilities of both classes and ROC curve. If target is non-binary it shows confusion matrix. File type: output. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/reference/classification/ref_output_plot_decision_tree.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 * **criterion** (*string*) - ("gini") The function to measure the quality of a split. Values: gini (for the Gini impurity), entropy (for the information gain).
35 * **max_depth** (*int*) - (4) [1~100|1] The maximum depth of the model. If None, then nodes are expanded until all leaves are pure or until all leaves contain less than min_samples_split samples.
36 * **normalize_cm** (*bool*) - (False) Whether or not to normalize the confusion matrix.
37 * **random_state_method** (*int*) - (5) [1~1000|1] Controls the randomness of the estimator.
38 * **random_state_train_test** (*int*) - (5) [1~1000|1] Controls the shuffling applied to the data before applying the split.
39 * **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.
40 * **scale** (*bool*) - (False) Whether or not to scale the input dataset.
41 * **remove_tmp** (*bool*) - (True) [WF property] Remove temporal files.
42 * **restart** (*bool*) - (False) [WF property] Do not execute if output files exist.
43 * **sandbox_path** (*str*) - ("./") [WF property] Parent path to the sandbox directory.
45 Examples:
46 This is a use example of how to use the building block from Python::
48 from biobb_ml.classification.decision_tree import decision_tree
49 prop = {
50 'independent_vars': {
51 'columns': [ 'column1', 'column2', 'column3' ]
52 },
53 'target': {
54 'column': 'target'
55 },
56 'criterion': 'entropy',
57 'test_size': 0.2
58 }
59 decision_tree(input_dataset_path='/path/to/myDataset.csv',
60 output_model_path='/path/to/newModel.pkl',
61 output_test_table_path='/path/to/newTable.csv',
62 output_plot_path='/path/to/newPlot.png',
63 properties=prop)
65 Info:
66 * wrapped_software:
67 * name: scikit-learn DecisionTreeClassifier
68 * version: >=0.24.2
69 * license: BSD 3-Clause
70 * ontology:
71 * name: EDAM
72 * schema: http://edamontology.org/EDAM.owl
74 """
76 def __init__(self, input_dataset_path, output_model_path,
77 output_test_table_path=None, output_plot_path=None, properties=None, **kwargs) -> None:
78 properties = properties or {}
80 # Call parent class constructor
81 super().__init__(properties)
82 self.locals_var_dict = locals().copy()
84 # Input/Output files
85 self.io_dict = {
86 "in": {"input_dataset_path": input_dataset_path},
87 "out": {"output_model_path": output_model_path, "output_test_table_path": output_test_table_path, "output_plot_path": output_plot_path}
88 }
90 # Properties specific for BB
91 self.independent_vars = properties.get('independent_vars', {})
92 self.target = properties.get('target', {})
93 self.weight = properties.get('weight', {})
94 self.criterion = properties.get('criterion', 'gini')
95 self.max_depth = properties.get('max_depth', 4)
96 self.normalize_cm = properties.get('normalize_cm', False)
97 self.random_state_method = properties.get('random_state_method', 5)
98 self.random_state_train_test = properties.get('random_state_train_test', 5)
99 self.test_size = properties.get('test_size', 0.2)
100 self.scale = properties.get('scale', False)
101 self.properties = properties
103 # Check the properties
104 self.check_properties(properties)
105 self.check_arguments()
107 def check_data_params(self, out_log, err_log):
108 """ Checks all the input/output paths and parameters """
109 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__)
110 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__)
111 if self.io_dict["out"]["output_test_table_path"]:
112 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__)
113 if self.io_dict["out"]["output_plot_path"]:
114 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__)
116 @launchlogger
117 def launch(self) -> int:
118 """Execute the :class:`DecisionTree <classification.decision_tree.DecisionTree>` classification.decision_tree.DecisionTree object."""
120 # check input/output paths and parameters
121 self.check_data_params(self.out_log, self.err_log)
123 # Setup Biobb
124 if self.check_restart():
125 return 0
126 self.stage_files()
128 # load dataset
129 fu.log('Getting dataset from %s' % self.io_dict["in"]["input_dataset_path"], self.out_log, self.global_log)
130 if 'columns' in self.independent_vars:
131 labels = getHeader(self.io_dict["in"]["input_dataset_path"])
132 skiprows = 1
133 else:
134 labels = None
135 skiprows = None
136 data = pd.read_csv(self.io_dict["in"]["input_dataset_path"], header=None, sep="\\s+|;|:|,|\t", engine="python", skiprows=skiprows, names=labels)
138 # declare inputs, targets and weights
139 # the inputs are all the independent variables
140 X = getIndependentVars(self.independent_vars, data, self.out_log, self.__class__.__name__)
141 fu.log('Independent variables: [%s]' % (getIndependentVarsList(self.independent_vars)), self.out_log, self.global_log)
142 # target
143 y = getTarget(self.target, data, self.out_log, self.__class__.__name__)
144 fu.log('Target: %s' % (getTargetValue(self.target)), self.out_log, self.global_log)
145 # weights
146 if self.weight:
147 w = getWeight(self.weight, data, self.out_log, self.__class__.__name__)
148 fu.log('Weight column provided', self.out_log, self.global_log)
150 # train / test split
151 fu.log('Creating train and test sets', self.out_log, self.global_log)
152 arrays_sets = (X, y)
153 # if user provide weights
154 if self.weight:
155 arrays_sets = arrays_sets + (w,)
156 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)
157 else:
158 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)
160 # scale dataset
161 if self.scale:
162 fu.log('Scaling dataset', self.out_log, self.global_log)
163 scaler = StandardScaler()
164 X_train = scaler.fit_transform(X_train)
166 # classification
167 fu.log('Training dataset applying decision tree classification', self.out_log, self.global_log)
168 model = DecisionTreeClassifier(criterion=self.criterion, max_depth=self.max_depth, random_state=self.random_state_method)
169 arrays_fit = (X_train, y_train)
170 # if user provide weights
171 if self.weight:
172 arrays_fit = arrays_fit + (w_train,)
174 model.fit(*arrays_fit)
176 y_hat_train = model.predict(X_train)
177 # classification report
178 cr_train = classification_report(y_train, y_hat_train)
179 # log loss
180 yhat_prob_train = model.predict_proba(X_train)
181 l_loss_train = log_loss(y_train, yhat_prob_train)
182 fu.log('Calculating scores and report for training dataset\n\nCLASSIFICATION REPORT\n\n%s\nLog loss: %.3f\n' % (cr_train, l_loss_train), self.out_log, self.global_log)
184 # compute confusion matrix
185 cnf_matrix_train = confusion_matrix(y_train, y_hat_train)
186 np.set_printoptions(precision=2)
187 if self.normalize_cm:
188 cnf_matrix_train = cnf_matrix_train.astype('float') / cnf_matrix_train.sum(axis=1)[:, np.newaxis]
189 cm_type = 'NORMALIZED CONFUSION MATRIX'
190 else:
191 cm_type = 'CONFUSION MATRIX, WITHOUT NORMALIZATION'
193 fu.log('Calculating confusion matrix for training dataset\n\n%s\n\n%s\n' % (cm_type, cnf_matrix_train), self.out_log, self.global_log)
195 # testing
196 # predict data from x_test
197 if self.scale:
198 X_test = scaler.transform(X_test)
199 y_hat_test = model.predict(X_test)
200 test_table = pd.DataFrame()
201 y_hat_prob = model.predict_proba(X_test)
202 y_hat_prob = np.around(y_hat_prob, decimals=2)
203 y_hat_prob = tuple(map(tuple, y_hat_prob))
204 test_table['P' + np.array2string(np.unique(y_test))] = y_hat_prob
205 y_test = y_test.reset_index(drop=True)
206 test_table['target'] = y_test
207 fu.log('Testing\n\nTEST DATA\n\n%s\n' % test_table, self.out_log, self.global_log)
209 # classification report
210 cr = classification_report(y_test, y_hat_test)
211 # log loss
212 yhat_prob = model.predict_proba(X_test)
213 l_loss = log_loss(y_test, yhat_prob)
214 fu.log('Calculating scores and report for testing dataset\n\nCLASSIFICATION REPORT\n\n%s\nLog loss: %.3f\n' % (cr, l_loss), self.out_log, self.global_log)
216 # compute confusion matrix
217 cnf_matrix = confusion_matrix(y_test, y_hat_test)
218 np.set_printoptions(precision=2)
219 if self.normalize_cm:
220 cnf_matrix = cnf_matrix.astype('float') / cnf_matrix.sum(axis=1)[:, np.newaxis]
221 cm_type = 'NORMALIZED CONFUSION MATRIX'
222 else:
223 cm_type = 'CONFUSION MATRIX, WITHOUT NORMALIZATION'
225 fu.log('Calculating confusion matrix for testing dataset\n\n%s\n\n%s\n' % (cm_type, cnf_matrix), self.out_log, self.global_log)
227 if (self.io_dict["out"]["output_test_table_path"]):
228 fu.log('Saving testing data to %s' % self.io_dict["out"]["output_test_table_path"], self.out_log, self.global_log)
229 test_table.to_csv(self.io_dict["out"]["output_test_table_path"], index=False, header=True)
231 # plot
232 if self.io_dict["out"]["output_plot_path"]:
233 vs = y.unique().tolist()
234 vs.sort()
235 if len(vs) > 2:
236 plot = plotMultipleCM(cnf_matrix_train, cnf_matrix, self.normalize_cm, vs)
237 fu.log('Saving confusion matrix plot to %s' % self.io_dict["out"]["output_plot_path"], self.out_log, self.global_log)
238 else:
239 plot = plotBinaryClassifier(model, yhat_prob_train, yhat_prob, cnf_matrix_train, cnf_matrix, y_train, y_test, normalize=self.normalize_cm)
240 fu.log('Saving binary classifier evaluator plot to %s' % self.io_dict["out"]["output_plot_path"], self.out_log, self.global_log)
241 plot.savefig(self.io_dict["out"]["output_plot_path"], dpi=150)
243 # save model, scaler and parameters
244 tv = y.unique().tolist()
245 tv.sort()
246 variables = {
247 'target': self.target,
248 'independent_vars': self.independent_vars,
249 'scale': self.scale,
250 'target_values': tv
251 }
252 fu.log('Saving model to %s' % self.io_dict["out"]["output_model_path"], self.out_log, self.global_log)
253 with open(self.io_dict["out"]["output_model_path"], "wb") as f:
254 joblib.dump(model, f)
255 if self.scale:
256 joblib.dump(scaler, f)
257 joblib.dump(variables, f)
259 # Copy files to host
260 self.copy_to_host()
262 self.tmp_files.extend([
263 self.stage_io_dict.get("unique_dir")
264 ])
265 self.remove_tmp_files()
267 self.check_arguments(output_files_created=True, raise_exception=False)
269 return 0
272def decision_tree(input_dataset_path: str, output_model_path: str, output_test_table_path: str = None, output_plot_path: str = None, properties: dict = None, **kwargs) -> int:
273 """Execute the :class:`DecisionTree <classification.decision_tree.DecisionTree>` class and
274 execute the :meth:`launch() <classification.decision_tree.DecisionTree.launch>` method."""
276 return DecisionTree(input_dataset_path=input_dataset_path,
277 output_model_path=output_model_path,
278 output_test_table_path=output_test_table_path,
279 output_plot_path=output_plot_path,
280 properties=properties, **kwargs).launch()
283def main():
284 """Command line execution of this building block. Please check the command line documentation."""
285 parser = argparse.ArgumentParser(description="Wrapper of the scikit-learn DecisionTreeClassifier method. ", formatter_class=lambda prog: argparse.RawTextHelpFormatter(prog, width=99999))
286 parser.add_argument('--config', required=False, help='Configuration file')
288 # Specific args of each building block
289 required_args = parser.add_argument_group('required arguments')
290 required_args.add_argument('--input_dataset_path', required=True, help='Path to the input dataset. Accepted formats: csv.')
291 required_args.add_argument('--output_model_path', required=True, help='Path to the output model file. Accepted formats: pkl.')
292 parser.add_argument('--output_test_table_path', required=False, help='Path to the test table file. Accepted formats: csv.')
293 parser.add_argument('--output_plot_path', required=False, help='Path to the statistics plot. If target is binary it shows confusion matrix, distributions of the predicted probabilities of both classes and ROC curve. If target is non-binary it shows confusion matrix. Accepted formats: png.')
295 args = parser.parse_args()
296 args.config = args.config or "{}"
297 properties = settings.ConfReader(config=args.config).get_prop_dic()
299 # Specific call of each building block
300 decision_tree(input_dataset_path=args.input_dataset_path,
301 output_model_path=args.output_model_path,
302 output_test_table_path=args.output_test_table_path,
303 output_plot_path=args.output_plot_path,
304 properties=properties)
307if __name__ == '__main__':
308 main()