sklearn_train_test_split: train_test

comparison train_test_eval.py @ 6:13b9ac5d277c draft

"planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/sklearn commit 208a8d348e7c7a182cfbe1b6f17868146428a7e2"

author	bgruening
date	Tue, 13 Apr 2021 22:24:07 +0000
parents	5a092779412e
children	3312fb686ffb

comparison

equal deleted inserted replaced

-:ce2fd1edbc6e
+:13b9ac5d277c
 import argparse
-import joblib
 import json
-import numpy as np
 import os
-import pandas as pd
 import pickle
 import warnings
 from itertools import chain
+import joblib
+import numpy as np
+import pandas as pd
+from galaxy_ml.model_validations import train_test_split
+from galaxy_ml.utils import (
+get_module,
+get_scoring,
+load_model,
+read_columns,
+SafeEval,
+try_get_attr,
+)
 from scipy.io import mmread
-from sklearn.base import clone
+from sklearn import pipeline
-from sklearn import (cluster, compose, decomposition, ensemble,
-feature_extraction, feature_selection,
-gaussian_process, kernel_approximation, metrics,
-model_selection, naive_bayes, neighbors,
-pipeline, preprocessing, svm, linear_model,
-tree, discriminant_analysis)
-from sklearn.exceptions import FitFailedWarning
 from sklearn.metrics.scorer import _check_multimetric_scoring
-from sklearn.model_selection._validation import _score, cross_validate
 from sklearn.model_selection import _search, _validation
+from sklearn.model_selection._validation import _score
 from sklearn.utils import indexable, safe_indexing
-from galaxy_ml.model_validations import train_test_split
-from galaxy_ml.utils import (SafeEval, get_scoring, load_model,
+_fit_and_score = try_get_attr("galaxy_ml.model_validations", "_fit_and_score")
-read_columns, try_get_attr, get_module)
+setattr(_search, "_fit_and_score", _fit_and_score)
+setattr(_validation, "_fit_and_score", _fit_and_score)
-_fit_and_score = try_get_attr('galaxy_ml.model_validations', '_fit_and_score')
+N_JOBS = int(os.environ.get("GALAXY_SLOTS", 1))
-setattr(_search, '_fit_and_score', _fit_and_score)
+CACHE_DIR = os.path.join(os.getcwd(), "cached")
-setattr(_validation, '_fit_and_score', _fit_and_score)
-N_JOBS = int(os.environ.get('GALAXY_SLOTS', 1))
-CACHE_DIR = os.path.join(os.getcwd(), 'cached')
 del os
-NON_SEARCHABLE = ('n_jobs', 'pre_dispatch', 'memory', '_path',
+NON_SEARCHABLE = ("n_jobs", "pre_dispatch", "memory", "_path", "nthread", "callbacks")
-'nthread', 'callbacks')
+ALLOWED_CALLBACKS = (
-ALLOWED_CALLBACKS = ('EarlyStopping', 'TerminateOnNaN', 'ReduceLROnPlateau',
+"EarlyStopping",
-'CSVLogger', 'None')
+"TerminateOnNaN",
+"ReduceLROnPlateau",
+"CSVLogger",
+"None",
+)
 def _eval_swap_params(params_builder):
 swap_params = {}
-for p in params_builder['param_set']:
+for p in params_builder["param_set"]:
-swap_value = p['sp_value'].strip()
+swap_value = p["sp_value"].strip()
-if swap_value == '':
+if swap_value == "":
 continue
-param_name = p['sp_name']
+param_name = p["sp_name"]
 if param_name.lower().endswith(NON_SEARCHABLE):
-warnings.warn("Warning: `%s` is not eligible for search and was "
+warnings.warn(
-"omitted!" % param_name)
+"Warning: `%s` is not eligible for search and was "
+"omitted!" % param_name
+)
 continue
-if not swap_value.startswith(':'):
+if not swap_value.startswith(":"):
 safe_eval = SafeEval(load_scipy=True, load_numpy=True)
 ev = safe_eval(swap_value)
 else:
 # Have `:` before search list, asks for estimator evaluatio
 safe_eval_es = SafeEval(load_estimators=True)
 if arr is None:
 nones.append(idx)
 else:
 new_arrays.append(arr)
-if kwargs['shuffle'] == 'None':
+if kwargs["shuffle"] == "None":
-kwargs['shuffle'] = None
+kwargs["shuffle"] = None
-group_names = kwargs.pop('group_names', None)
+group_names = kwargs.pop("group_names", None)
 if group_names is not None and group_names.strip():
-group_names = [name.strip() for name in
+group_names = [name.strip() for name in group_names.split(",")]
-group_names.split(',')]
 new_arrays = indexable(*new_arrays)
-groups = kwargs['labels']
+groups = kwargs["labels"]
 n_samples = new_arrays[0].shape[0]
 index_arr = np.arange(n_samples)
 test = index_arr[np.isin(groups, group_names)]
 train = index_arr[~np.isin(groups, group_names)]
-rval = list(chain.from_iterable(
+rval = list(
-(safe_indexing(a, train),
+chain.from_iterable(
-safe_indexing(a, test)) for a in new_arrays))
+(safe_indexing(a, train), safe_indexing(a, test)) for a in new_arrays
+)
+)
 else:
 rval = train_test_split(*new_arrays, **kwargs)
 for pos in nones:
 rval[pos * 2: 2] = [None, None]
 return rval
-def main(inputs, infile_estimator, infile1, infile2,
+def main(
-outfile_result, outfile_object=None,
+inputs,
-outfile_weights=None, groups=None,
+infile_estimator,
-ref_seq=None, intervals=None, targets=None,
+infile1,
-fasta_path=None):
+infile2,
+outfile_result,
+outfile_object=None,
+outfile_weights=None,
+groups=None,
+ref_seq=None,
+intervals=None,
+targets=None,
+fasta_path=None,
+):
 """
 Parameter
 ---------
 inputs : str
 File path to galaxy tool parameter
 File path to dataset compressed target bed file
 fasta_path : str
 File path to dataset containing fasta file
 """
-warnings.simplefilter('ignore')
+warnings.simplefilter("ignore")
-with open(inputs, 'r') as param_handler:
+with open(inputs, "r") as param_handler:
 params = json.load(param_handler)
 #  load estimator
-with open(infile_estimator, 'rb') as estimator_handler:
+with open(infile_estimator, "rb") as estimator_handler:
 estimator = load_model(estimator_handler)
 # swap hyperparameter
-swapping = params['experiment_schemes']['hyperparams_swapping']
+swapping = params["experiment_schemes"]["hyperparams_swapping"]
 swap_params = _eval_swap_params(swapping)
 estimator.set_params(**swap_params)
 estimator_params = estimator.get_params()
 # store read dataframe object
 loaded_df = {}
-input_type = params['input_options']['selected_input']
+input_type = params["input_options"]["selected_input"]
 # tabular input
-if input_type == 'tabular':
+if input_type == "tabular":
-header = 'infer' if params['input_options']['header1'] else None
+header = "infer" if params["input_options"]["header1"] else None
-column_option = (params['input_options']['column_selector_options_1']
+column_option = params["input_options"]["column_selector_options_1"][
-['selected_column_selector_option'])
+"selected_column_selector_option"
-if column_option in ['by_index_number', 'all_but_by_index_number',
+]
-'by_header_name', 'all_but_by_header_name']:
+if column_option in [
-c = params['input_options']['column_selector_options_1']['col1']
+"by_index_number",
+"all_but_by_index_number",
+"by_header_name",
+"all_but_by_header_name",
+]:
+c = params["input_options"]["column_selector_options_1"]["col1"]
 else:
 c = None
 df_key = infile1 + repr(header)
-df = pd.read_csv(infile1, sep='\t', header=header,
+df = pd.read_csv(infile1, sep="\t", header=header, parse_dates=True)
-parse_dates=True)
 loaded_df[df_key] = df
 X = read_columns(df, c=c, c_option=column_option).astype(float)
 # sparse input
-elif input_type == 'sparse':
+elif input_type == "sparse":
-X = mmread(open(infile1, 'r'))
+X = mmread(open(infile1, "r"))
 # fasta_file input
-elif input_type == 'seq_fasta':
+elif input_type == "seq_fasta":
-pyfaidx = get_module('pyfaidx')
+pyfaidx = get_module("pyfaidx")
 sequences = pyfaidx.Fasta(fasta_path)
 n_seqs = len(sequences.keys())
 X = np.arange(n_seqs)[:, np.newaxis]
 for param in estimator_params.keys():
-if param.endswith('fasta_path'):
+if param.endswith("fasta_path"):
-estimator.set_params(
+estimator.set_params(**{param: fasta_path})
-**{param: fasta_path})
 break
 else:
 raise ValueError(
 "The selected estimator doesn't support "
 "fasta file input! Please consider using "
 "KerasGBatchClassifier with "
 "FastaDNABatchGenerator/FastaProteinBatchGenerator "
 "or having GenomeOneHotEncoder/ProteinOneHotEncoder "
-"in pipeline!")
+"in pipeline!"
+)
-elif input_type == 'refseq_and_interval':
+elif input_type == "refseq_and_interval":
 path_params = {
-'data_batch_generator__ref_genome_path': ref_seq,
+"data_batch_generator__ref_genome_path": ref_seq,
-'data_batch_generator__intervals_path': intervals,
+"data_batch_generator__intervals_path": intervals,
-'data_batch_generator__target_path': targets
+"data_batch_generator__target_path": targets,
 }
 estimator.set_params(**path_params)
 n_intervals = sum(1 for line in open(intervals))
 X = np.arange(n_intervals)[:, np.newaxis]
 # Get target y
-header = 'infer' if params['input_options']['header2'] else None
+header = "infer" if params["input_options"]["header2"] else None
-column_option = (params['input_options']['column_selector_options_2']
+column_option = params["input_options"]["column_selector_options_2"][
-['selected_column_selector_option2'])
+"selected_column_selector_option2"
-if column_option in ['by_index_number', 'all_but_by_index_number',
+]
-'by_header_name', 'all_but_by_header_name']:
+if column_option in [
-c = params['input_options']['column_selector_options_2']['col2']
+"by_index_number",
+"all_but_by_index_number",
+"by_header_name",
+"all_but_by_header_name",
+]:
+c = params["input_options"]["column_selector_options_2"]["col2"]
 else:
 c = None
 df_key = infile2 + repr(header)
 if df_key in loaded_df:
 infile2 = loaded_df[df_key]
 else:
-infile2 = pd.read_csv(infile2, sep='\t',
+infile2 = pd.read_csv(infile2, sep="\t", header=header, parse_dates=True)
-header=header, parse_dates=True)
 loaded_df[df_key] = infile2
-y = read_columns(
+y = read_columns(infile2,
-infile2,
+c=c,
-c=c,
+c_option=column_option,
-c_option=column_option,
+sep='\t',
-sep='\t',
+header=header,
-header=header,
+parse_dates=True)
-parse_dates=True)
 if len(y.shape) == 2 and y.shape[1] == 1:
 y = y.ravel()
-if input_type == 'refseq_and_interval':
+if input_type == "refseq_and_interval":
-estimator.set_params(
+estimator.set_params(data_batch_generator__features=y.ravel().tolist())
-data_batch_generator__features=y.ravel().tolist())
 y = None
 # end y
 # load groups
 if groups:
-groups_selector = (params['experiment_schemes']['test_split']
+groups_selector = (
-['split_algos']).pop('groups_selector')
+params["experiment_schemes"]["test_split"]["split_algos"]
+).pop("groups_selector")
-header = 'infer' if groups_selector['header_g'] else None
-column_option = \
+header = "infer" if groups_selector["header_g"] else None
-(groups_selector['column_selector_options_g']
+column_option = groups_selector["column_selector_options_g"][
-['selected_column_selector_option_g'])
+"selected_column_selector_option_g"
-if column_option in ['by_index_number', 'all_but_by_index_number',
+]
-'by_header_name', 'all_but_by_header_name']:
+if column_option in [
-c = groups_selector['column_selector_options_g']['col_g']
+"by_index_number",
+"all_but_by_index_number",
+"by_header_name",
+"all_but_by_header_name",
+]:
+c = groups_selector["column_selector_options_g"]["col_g"]
 else:
 c = None
 df_key = groups + repr(header)
 if df_key in loaded_df:
 groups = loaded_df[df_key]
-groups = read_columns(
+groups = read_columns(groups,
-groups,
+c=c,
-c=c,
+c_option=column_option,
-c_option=column_option,
+sep='\t',
-sep='\t',
+header=header,
-header=header,
+parse_dates=True)
-parse_dates=True)
 groups = groups.ravel()
 # del loaded_df
 del loaded_df
 # handle memory
 memory = joblib.Memory(location=CACHE_DIR, verbose=0)
 # cache iraps_core fits could increase search speed significantly
-if estimator.__class__.__name__ == 'IRAPSClassifier':
+if estimator.__class__.__name__ == "IRAPSClassifier":
 estimator.set_params(memory=memory)
 else:
 # For iraps buried in pipeline
 new_params = {}
 for p, v in estimator_params.items():
-if p.endswith('memory'):
+if p.endswith("memory"):
 # for case of `__irapsclassifier__memory`
-if len(p) > 8 and p[:-8].endswith('irapsclassifier'):
+if len(p) > 8 and p[:-8].endswith("irapsclassifier"):
 # cache iraps_core fits could increase search
 # speed significantly
 new_params[p] = memory
 # security reason, we don't want memory being
 # modified unexpectedly
 elif v:
 new_params[p] = None
 # handle n_jobs
-elif p.endswith('n_jobs'):
+elif p.endswith("n_jobs"):
 # For now, 1 CPU is suggested for iprasclassifier
-if len(p) > 8 and p[:-8].endswith('irapsclassifier'):
+if len(p) > 8 and p[:-8].endswith("irapsclassifier"):
 new_params[p] = 1
 else:
 new_params[p] = N_JOBS
 # for security reason, types of callback are limited
-elif p.endswith('callbacks'):
+elif p.endswith("callbacks"):
 for cb in v:
-cb_type = cb['callback_selection']['callback_type']
+cb_type = cb["callback_selection"]["callback_type"]
 if cb_type not in ALLOWED_CALLBACKS:
-raise ValueError(
+raise ValueError("Prohibited callback type: %s!" % cb_type)
-"Prohibited callback type: %s!" % cb_type)
 estimator.set_params(**new_params)
 # handle scorer, convert to scorer dict
-scoring = params['experiment_schemes']['metrics']['scoring']
+# Check if scoring is specified
+scoring = params["experiment_schemes"]["metrics"].get("scoring", None)
+if scoring is not None:
+# get_scoring() expects secondary_scoring to be a comma separated string (not a list)
+# Check if secondary_scoring is specified
+secondary_scoring = scoring.get("secondary_scoring", None)
+if secondary_scoring is not None:
+# If secondary_scoring is specified, convert the list into comman separated string
+scoring["secondary_scoring"] = ",".join(scoring["secondary_scoring"])
 scorer = get_scoring(scoring)
 scorer, _ = _check_multimetric_scoring(estimator, scoring=scorer)
 # handle test (first) split
-test_split_options = (params['experiment_schemes']
+test_split_options = params["experiment_schemes"]["test_split"]["split_algos"]
-['test_split']['split_algos'])
+if test_split_options["shuffle"] == "group":
-if test_split_options['shuffle'] == 'group':
+test_split_options["labels"] = groups
-test_split_options['labels'] = groups
+if test_split_options["shuffle"] == "stratified":
-if test_split_options['shuffle'] == 'stratified':
 if y is not None:
-test_split_options['labels'] = y
+test_split_options["labels"] = y
 else:
-raise ValueError("Stratified shuffle split is not "
+raise ValueError(
-"applicable on empty target values!")
+"Stratified shuffle split is not " "applicable on empty target values!"
+)
-X_train, X_test, y_train, y_test, groups_train, groups_test = \
-train_test_split_none(X, y, groups, **test_split_options)
+X_train, X_test, y_train, y_test, groups_train, _groups_test = train_test_split_none(
+X, y, groups, **test_split_options
-exp_scheme = params['experiment_schemes']['selected_exp_scheme']
+)
+exp_scheme = params["experiment_schemes"]["selected_exp_scheme"]
 # handle validation (second) split
-if exp_scheme == 'train_val_test':
+if exp_scheme == "train_val_test":
-val_split_options = (params['experiment_schemes']
+val_split_options = params["experiment_schemes"]["val_split"]["split_algos"]
-['val_split']['split_algos'])
+if val_split_options["shuffle"] == "group":
-if val_split_options['shuffle'] == 'group':
+val_split_options["labels"] = groups_train
-val_split_options['labels'] = groups_train
+if val_split_options["shuffle"] == "stratified":
-if val_split_options['shuffle'] == 'stratified':
 if y_train is not None:
-val_split_options['labels'] = y_train
+val_split_options["labels"] = y_train
 else:
-raise ValueError("Stratified shuffle split is not "
+raise ValueError(
-"applicable on empty target values!")
+"Stratified shuffle split is not "
+"applicable on empty target values!"
-X_train, X_val, y_train, y_val, groups_train, groups_val = \
+)
-train_test_split_none(X_train, y_train, groups_train,
-**val_split_options)
+(
+X_train,
+X_val,
+y_train,
+y_val,
+groups_train,
+_groups_val,
+) = train_test_split_none(X_train, y_train, groups_train, **val_split_options)
 # train and eval
-if hasattr(estimator, 'validation_data'):
+if hasattr(estimator, "validation_data"):
-if exp_scheme == 'train_val_test':
+if exp_scheme == "train_val_test":
-estimator.fit(X_train, y_train,
+estimator.fit(X_train, y_train, validation_data=(X_val, y_val))
-validation_data=(X_val, y_val))
+else:
-else:
+estimator.fit(X_train, y_train, validation_data=(X_test, y_test))
-estimator.fit(X_train, y_train,
-validation_data=(X_test, y_test))
 else:
 estimator.fit(X_train, y_train)
-if hasattr(estimator, 'evaluate'):
+if hasattr(estimator, "evaluate"):
-scores = estimator.evaluate(X_test, y_test=y_test,
+scores = estimator.evaluate(
-scorer=scorer,
+X_test, y_test=y_test, scorer=scorer, is_multimetric=True
-is_multimetric=True)
+)
 else:
-scores = _score(estimator, X_test, y_test, scorer,
+scores = _score(estimator, X_test, y_test, scorer, is_multimetric=True)
-is_multimetric=True)
 # handle output
 for name, score in scores.items():
 scores[name] = [score]
 df = pd.DataFrame(scores)
 df = df[sorted(df.columns)]
-df.to_csv(path_or_buf=outfile_result, sep='\t',
+df.to_csv(path_or_buf=outfile_result, sep="\t", header=True, index=False)
-header=True, index=False)
 memory.clear(warn=False)
 if outfile_object:
 main_est = estimator
 if isinstance(estimator, pipeline.Pipeline):
 main_est = estimator.steps[-1][-1]
-if hasattr(main_est, 'model_') \
+if hasattr(main_est, "model_") and hasattr(main_est, "save_weights"):
-and hasattr(main_est, 'save_weights'):
 if outfile_weights:
 main_est.save_weights(outfile_weights)
-del main_est.model_
+if getattr(main_est, "model_", None):
-del main_est.fit_params
+del main_est.model_
-del main_est.model_class_
+if getattr(main_est, "fit_params", None):
-del main_est.validation_data
+del main_est.fit_params
-if getattr(main_est, 'data_generator_', None):
+if getattr(main_est, "model_class_", None):
+del main_est.model_class_
+if getattr(main_est, "validation_data", None):
+del main_est.validation_data
+if getattr(main_est, "data_generator_", None):
 del main_est.data_generator_
-with open(outfile_object, 'wb') as output_handler:
+with open(outfile_object, "wb") as output_handler:
-pickle.dump(estimator, output_handler,
+pickle.dump(estimator, output_handler, pickle.HIGHEST_PROTOCOL)
-pickle.HIGHEST_PROTOCOL)
+if __name__ == "__main__":
-if __name__ == '__main__':
 aparser = argparse.ArgumentParser()
 aparser.add_argument("-i", "--inputs", dest="inputs", required=True)
 aparser.add_argument("-e", "--estimator", dest="infile_estimator")
 aparser.add_argument("-X", "--infile1", dest="infile1")
 aparser.add_argument("-y", "--infile2", dest="infile2")
 aparser.add_argument("-b", "--intervals", dest="intervals")
 aparser.add_argument("-t", "--targets", dest="targets")
 aparser.add_argument("-f", "--fasta_path", dest="fasta_path")
 args = aparser.parse_args()
-main(args.inputs, args.infile_estimator, args.infile1, args.infile2,
+main(
-args.outfile_result, outfile_object=args.outfile_object,
+args.inputs,
-outfile_weights=args.outfile_weights, groups=args.groups,
+args.infile_estimator,
-ref_seq=args.ref_seq, intervals=args.intervals,
+args.infile1,
-targets=args.targets, fasta_path=args.fasta_path)
+args.infile2,
+args.outfile_result,
+outfile_object=args.outfile_object,
+outfile_weights=args.outfile_weights,
+groups=args.groups,
+ref_seq=args.ref_seq,
+intervals=args.intervals,
+targets=args.targets,
+fasta_path=args.fasta_path,
+)

Mercurial > repos > bgruening > sklearn_train_test_split

comparison train_test_eval.py @ 6:13b9ac5d277c draft