Update bayesclass models

Add excel output of reports of be_list

.github/workflows/main.yml

@@ -12,11 +12,8 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [macos-latest, ubuntu-latest]
-        python: ["3.10", "3.11"]
-        exclude:
-          - os: macos-latest
-            python: "3.11"
+        os: [ubuntu-latest]
+        python: ["3.10"]
 
     steps:
       - uses: actions/checkout@v3
@@ -46,6 +43,7 @@ jobs:
           pip install -q --upgrade pip
           pip install -q -r requirements.txt
           pip install -q --upgrade codecov coverage black flake8
+          git clone https://github.com/Doctorado-ML/bayesclass.git
       - name: Lint
         run: |
           black --check --diff benchmark

benchmark/Arguments.py

@@ -36,6 +36,7 @@ class EnvDefault(argparse.Action):
         self, envvar, required=True, default=None, mandatory=False, **kwargs
     ):
         self._args = EnvData.load()
+        self._overrides = {}
         if required and not mandatory:
             default = self._args[envvar]
             required = False
@@ -47,24 +48,27 @@ class EnvDefault(argparse.Action):
         setattr(namespace, self.dest, values)
 
 
-class Arguments:
-    def __init__(self):
-        self.ap = argparse.ArgumentParser()
+class Arguments(argparse.ArgumentParser):
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
         models_data = Models.define_models(random_state=0)
+        self._overrides = {}
+        self._subparser = None
         self.parameters = {
-            "best": [
-                ("-b", "--best"),
+            "best_paramfile": [
+                ("-b", "--best_paramfile"),
                 {
-                    "type": str,
+                    "action": "store_true",
                     "required": False,
-                    "help": "best results of models",
+                    "default": False,
+                    "help": "Use best hyperparams file?",
                 },
             ],
             "color": [
                 ("-c", "--color"),
                 {
-                    "type": bool,
                     "required": False,
+                    "action": "store_true",
                     "default": False,
                     "help": "use colors for the tree",
                 },
@@ -72,8 +76,9 @@ class Arguments:
             "compare": [
                 ("-c", "--compare"),
                 {
-                    "type": bool,
+                    "action": "store_true",
                     "required": False,
+                    "default": False,
                     "help": "Compare accuracy with best results",
                 },
             ],
@@ -81,6 +86,8 @@ class Arguments:
                 ("-d", "--dataset"),
                 {
                     "type": str,
+                    "envvar": "dataset",  # for compatiblity with EnvDefault
+                    "action": EnvDefault,
                     "required": False,
                     "help": "dataset to work with",
                 },
@@ -88,38 +95,26 @@ class Arguments:
             "excel": [
                 ("-x", "--excel"),
                 {
-                    "type": bool,
                     "required": False,
+                    "action": "store_true",
                     "default": False,
                     "help": "Generate Excel File",
                 },
             ],
-            "file": [
-                ("-f", "--file"),
-                {"type": str, "required": False, "help": "Result file"},
-            ],
-            "grid": [
-                ("-g", "--grid"),
-                {
-                    "type": str,
-                    "required": False,
-                    "help": "grid results of model",
-                },
-            ],
             "grid_paramfile": [
                 ("-g", "--grid_paramfile"),
                 {
-                    "type": bool,
                     "required": False,
+                    "action": "store_true",
                     "default": False,
-                    "help": "Use best hyperparams file?",
+                    "help": "Use grid output hyperparams file?",
                 },
             ],
             "hidden": [
                 ("--hidden",),
                 {
-                    "type": str,
                     "required": False,
+                    "action": "store_true",
                     "default": False,
                     "help": "Show hidden results",
                 },
@@ -140,8 +135,8 @@ class Arguments:
             "lose": [
                 ("-l", "--lose"),
                 {
-                    "type": bool,
                     "default": False,
+                    "action": "store_true",
                     "required": False,
                     "help": "show lose results",
                 },
@@ -178,9 +173,10 @@ class Arguments:
             "nan": [
                 ("--nan",),
                 {
-                    "type": bool,
+                    "action": "store_true",
                     "required": False,
-                    "help": "Move nan results to hidden folder",
+                    "default": False,
+                    "help": "List nan results to hidden folder",
                 },
             ],
             "number": [
@@ -202,15 +198,6 @@ class Arguments:
                     "help": "number of folds",
                 },
             ],
-            "paramfile": [
-                ("-f", "--paramfile"),
-                {
-                    "type": bool,
-                    "required": False,
-                    "default": False,
-                    "help": "Use best hyperparams file?",
-                },
-            ],
             "platform": [
                 ("-P", "--platform"),
                 {
@@ -224,7 +211,7 @@ class Arguments:
             "quiet": [
                 ("-q", "--quiet"),
                 {
-                    "type": bool,
+                    "action": "store_true",
                     "required": False,
                     "default": False,
                 },
@@ -232,7 +219,7 @@ class Arguments:
             "report": [
                 ("-r", "--report"),
                 {
-                    "type": bool,
+                    "action": "store_true",
                     "default": False,
                     "required": False,
                     "help": "Report results",
@@ -250,14 +237,18 @@ class Arguments:
             ],
             "sql": [
                 ("-q", "--sql"),
-                {"type": bool, "required": False, "help": "Generate SQL File"},
+                {
+                    "required": False,
+                    "action": "store_true",
+                    "default": False,
+                    "help": "Generate SQL File",
+                },
             ],
             "stratified": [
                 ("-t", "--stratified"),
                 {
                     "action": EnvDefault,
                     "envvar": "stratified",
-                    "type": str,
                     "required": True,
                     "help": "Stratified",
                 },
@@ -265,8 +256,8 @@ class Arguments:
             "tex_output": [
                 ("-t", "--tex-output"),
                 {
-                    "type": bool,
                     "required": False,
+                    "action": "store_true",
                     "default": False,
                     "help": "Generate Tex file with the table",
                 },
@@ -278,8 +269,8 @@ class Arguments:
             "win": [
                 ("-w", "--win"),
                 {
-                    "type": bool,
                     "default": False,
+                    "action": "store_true",
                     "required": False,
                     "help": "show win results",
                 },
@@ -287,12 +278,43 @@ class Arguments:
         }
 
     def xset(self, *arg_name, **kwargs):
-        names, default = self.parameters[arg_name[0]]
-        self.ap.add_argument(
+        names, parameters = self.parameters[arg_name[0]]
+        if "overrides" in kwargs:
+            self._overrides[names[0]] = (kwargs["overrides"], kwargs["const"])
+            del kwargs["overrides"]
+        self.add_argument(
             *names,
-            **{**default, **kwargs},
+            **{**parameters, **kwargs},
         )
         return self
 
+    def add_subparser(
+        self, dest="subcommand", help_text="help for subcommand"
+    ):
+        self._subparser = self.add_subparsers(dest=dest, help=help_text)
+
+    def add_subparsers_options(self, subparser, arguments):
+        command, help_text = subparser
+        parser = self._subparser.add_parser(command, help=help_text)
+        for name, args in arguments:
+            try:
+                names, parameters = self.parameters[name]
+            except KeyError:
+                names = (name,)
+                parameters = {}
+            # Order of args is important
+            parser.add_argument(*names, **{**args, **parameters})
+
+    def add_exclusive(self, hyperparameters, required=False):
+        group = self.add_mutually_exclusive_group(required=required)
+        for name in hyperparameters:
+            names, parameters = self.parameters[name]
+            group.add_argument(*names, **parameters)
+
     def parse(self, args=None):
-        return self.ap.parse_args(args)
+        for key, (dest_key, value) in self._overrides.items():
+            if args is None:
+                args = sys.argv[1:]
+            if key in args:
+                args.extend((f"--{dest_key}", value))
+        return super().parse_args(args)

benchmark/Datasets.py

@@ -1,4 +1,5 @@
 import os
+from types import SimpleNamespace
 import pandas as pd
 import numpy as np
 from scipy.io import arff
@@ -31,6 +32,7 @@ class DatasetsArff:
         data = arff.loadarff(file_name)
         df = pd.DataFrame(data[0])
         df.dropna(axis=0, how="any", inplace=True)
+        self.dataset = df
         X = df.drop(class_name, axis=1)
         self.features = X.columns
         self.class_name = class_name
@@ -55,8 +57,12 @@ class DatasetsTanveer:
             sep="\t",
             index_col=0,
         )
-        X = data.drop("clase", axis=1).to_numpy()
+        X = data.drop("clase", axis=1)
+        self.features = X.columns
+        X = X.to_numpy()
         y = data["clase"].to_numpy()
+        self.dataset = data
+        self.class_name = "clase"
         return X, y
 
 
@@ -77,8 +83,11 @@ class DatasetsSurcov:
         )
         data.dropna(axis=0, how="any", inplace=True)
         self.columns = data.columns
-        col_list = ["class"]
-        X = data.drop(col_list, axis=1).to_numpy()
+        X = data.drop(["class"], axis=1)
+        self.features = X.columns
+        self.class_name = "class"
+        self.dataset = data
+        X = X.to_numpy()
         y = data["class"].to_numpy()
         return X, y
 
@@ -86,50 +95,48 @@ class DatasetsSurcov:
 class Datasets:
     def __init__(self, dataset_name=None):
         envData = EnvData.load()
-        class_name = getattr(
+        # DatasetsSurcov, DatasetsTanveer, DatasetsArff,...
+        source_name = getattr(
             __import__(__name__),
             f"Datasets{envData['source_data']}",
         )
-        self.load = (
-            self.load_discretized
-            if envData["discretize"] == "1"
-            else self.load_continuous
-        )
-        self.dataset = class_name()
+        self.discretize = envData["discretize"] == "1"
+        self.dataset = source_name()
         self.class_names = []
-        self._load_names()
-        if dataset_name is not None:
-            try:
-                class_name = self.class_names[
-                    self.data_sets.index(dataset_name)
-                ]
-                self.class_names = [class_name]
-            except ValueError:
-                raise ValueError(f"Unknown dataset: {dataset_name}")
-            self.data_sets = [dataset_name]
+        self.data_sets = []
+        # initialize self.class_names & self.data_sets
+        class_names, sets = self._init_names(dataset_name)
+        self.class_names = class_names
+        self.data_sets = sets
 
-    def _load_names(self):
+    def _init_names(self, dataset_name):
         file_name = os.path.join(self.dataset.folder(), Files.index)
         default_class = "class"
         with open(file_name) as f:
-            self.data_sets = f.read().splitlines()
-            self.class_names = [default_class] * len(self.data_sets)
-        if "," in self.data_sets[0]:
+            sets = f.read().splitlines()
+            class_names = [default_class] * len(sets)
+        if "," in sets[0]:
             result = []
             class_names = []
-            for data in self.data_sets:
+            for data in sets:
                 name, class_name = data.split(",")
                 result.append(name)
                 class_names.append(class_name)
-            self.data_sets = result
-            self.class_names = class_names
+            sets = result
+        # Set as dataset list the dataset passed as argument
+        if dataset_name is None:
+            return class_names, sets
+        try:
+            class_name = class_names[sets.index(dataset_name)]
+        except ValueError:
+            raise ValueError(f"Unknown dataset: {dataset_name}")
+        return [class_name], [dataset_name]
 
     def get_attributes(self, name):
-        class Attributes:
-            pass
-
-        X, y = self.load_continuous(name)
-        attr = Attributes()
+        tmp = self.discretize
+        self.discretize = False
+        X, y = self.load(name)
+        attr = SimpleNamespace()
         values, counts = np.unique(y, return_counts=True)
         comp = ""
         sep = ""
@@ -140,6 +147,7 @@ class Datasets:
         attr.classes = len(np.unique(y))
         attr.samples = X.shape[0]
         attr.features = X.shape[1]
+        self.discretize = tmp
         return attr
 
     def get_features(self):
@@ -148,14 +156,25 @@ class Datasets:
     def get_class_name(self):
         return self.dataset.class_name
 
-    def load_continuous(self, name):
+    def get_dataset(self):
+        return self.dataset.dataset
+
+    def load(self, name, dataframe=False):
         try:
             class_name = self.class_names[self.data_sets.index(name)]
-            return self.dataset.load(name, class_name)
+            X, y = self.dataset.load(name, class_name)
+            if self.discretize:
+                X = self.discretize_dataset(X, y)
+                dataset = pd.DataFrame(X, columns=self.get_features())
+                dataset[self.get_class_name()] = y
+                self.dataset.dataset = dataset
+            if dataframe:
+                return self.get_dataset()
+            return X, y
         except (ValueError, FileNotFoundError):
             raise ValueError(f"Unknown dataset: {name}")
 
-    def discretize(self, X, y):
+    def discretize_dataset(self, X, y):
         """Supervised discretization with Fayyad and Irani's MDLP algorithm.
 
         Parameters
@@ -173,14 +192,5 @@ class Datasets:
         Xdisc = discretiz.fit_transform(X, y)
         return Xdisc
 
-    def load_discretized(self, name, dataframe=False):
-        X, yd = self.load_continuous(name)
-        Xd = self.discretize(X, yd)
-        dataset = pd.DataFrame(Xd, columns=self.get_features())
-        dataset[self.get_class_name()] = yd
-        if dataframe:
-            return dataset
-        return Xd, yd
-
     def __iter__(self) -> Diterator:
         return Diterator(self.data_sets)

benchmark/Models.py

@@ -8,7 +8,7 @@ from sklearn.ensemble import (
 )
 from sklearn.svm import SVC
 from stree import Stree
-from bayesclass import TAN, KDB, AODE
+from bayesclass.clfs import TAN, KDB, AODE
 from wodt import Wodt
 from odte import Odte
 from xgboost import XGBClassifier

benchmark/Results.py

@@ -1,6 +1,7 @@
 import os
 import sys
 from operator import itemgetter
+from types import SimpleNamespace
 import math
 import json
 import abc
@@ -22,8 +23,8 @@ from .Utils import (
 from ._version import __version__
 
 
-def get_input(is_test):
-    return "test" if is_test else input()
+def get_input(message="", is_test=False):
+    return "test" if is_test else input(message)
 
 
 class BestResultsEver:
@@ -251,7 +252,7 @@ class ReportBest(BaseReport):
         "Hyperparameters",
     ]
 
-    def __init__(self, score, model, best, grid):
+    def __init__(self, score, model, best):
         name = (
             Files.best_results(score, model)
             if best
@@ -259,7 +260,6 @@ class ReportBest(BaseReport):
         )
         file_name = os.path.join(Folders.results, name)
         self.best = best
-        self.grid = grid
         self.score_name = score
         self.model = model
         super().__init__(file_name, best_file=True)
@@ -1353,13 +1353,14 @@ class StubReport(BaseReport):
 
 
 class Summary:
-    def __init__(self, hidden=False) -> None:
+    def __init__(self, hidden=False, compare=False) -> None:
         self.results = Files().get_all_results(hidden=hidden)
         self.data = []
         self.data_filtered = []
         self.datasets = {}
         self.models = set()
         self.hidden = hidden
+        self.compare = compare
 
     def get_models(self):
         return sorted(self.models)
@@ -1402,18 +1403,15 @@ class Summary:
                 self.data.append(entry)
 
     def get_results_criteria(
-        self,
-        score,
-        model,
-        input_data,
-        sort_key,
-        number,
+        self, score, model, input_data, sort_key, number, nan=False
     ):
         data = self.data.copy() if input_data is None else input_data
         if score:
             data = [x for x in data if x["score"] == score]
         if model:
             data = [x for x in data if x["model"] == model]
+        if nan:
+            data = [x for x in data if x["metric"] != x["metric"]]
         keys = (
             itemgetter(sort_key, "time")
             if sort_key == "date"
@@ -1431,11 +1429,12 @@ class Summary:
         input_data=None,
         sort_key="date",
         number=0,
+        nan=False,
     ) -> None:
         """Print the list of results"""
         if self.data_filtered == []:
             self.data_filtered = self.get_results_criteria(
-                score, model, input_data, sort_key, number
+                score, model, input_data, sort_key, number, nan=nan
             )
         if self.data_filtered == []:
             raise ValueError(NO_RESULTS)
@@ -1477,44 +1476,99 @@ class Summary:
             )
         )
 
-    def manage_results(self, excel, is_test):
+    def manage_results(self):
         """Manage results showed in the summary
         return True if excel file is created False otherwise
         """
-        num = ""
-        book = None
-        while True:
-            print(
-                "Which result do you want to report? (q to quit, r to list "
-                "again, number to report): ",
-                end="",
+
+        def process_file(num, command, path):
+            num = int(num)
+            name = self.data_filtered[num]["file"]
+            file_name_result = os.path.join(path, name)
+            verb1, verb2 = (
+                ("delete", "Deleting")
+                if command == cmd.delete
+                else (
+                    "hide",
+                    "Hiding",
+                )
             )
-            num = get_input(is_test)
-            if num == "r":
+            conf_message = (
+                TextColor.RED
+                + f"Are you sure to {verb1} {file_name_result} (y/n)? "
+            )
+            confirm = get_input(message=conf_message)
+            if confirm == "y":
+                print(TextColor.YELLOW + f"{verb2} {file_name_result}")
+                if command == cmd.delete:
+                    os.unlink(file_name_result)
+                else:
+                    os.rename(
+                        os.path.join(Folders.results, name),
+                        os.path.join(Folders.hidden_results, name),
+                    )
+                self.data_filtered.pop(num)
+                get_input(message="Press enter to continue")
                 self.list_results()
-            if num == "q":
-                if excel:
+
+        cmd = SimpleNamespace(
+            quit="q", relist="r", delete="d", hide="h", excel="e"
+        )
+        message = (
+            TextColor.ENDC
+            + f"Choose option {str(cmd).replace('namespace', '')}: "
+        )
+        path = Folders.hidden_results if self.hidden else Folders.results
+        book = None
+        max_value = len(self.data)
+        while True:
+            match get_input(message=message).split():
+                case [cmd.relist]:
+                    self.list_results()
+                case [cmd.quit]:
                     if book is not None:
                         book.close()
                         return True
-                return False
-            if num.isdigit() and int(num) < len(self.data) and int(num) >= 0:
-                rep = Report(self.data_filtered[int(num)]["file"], self.hidden)
-                rep.report()
-                if excel and not self.hidden:
+                    return False
+                case [cmd.hide, num] if num.isdigit() and int(num) < max_value:
+                    if self.hidden:
+                        print("Already hidden")
+                    else:
+                        process_file(num, path=path, command=cmd.hide)
+                case [cmd.delete, num] if num.isdigit() and int(
+                    num
+                ) < max_value:
+                    process_file(num=num, path=path, command=cmd.delete)
+                case [cmd.excel, num] if num.isdigit() and int(
+                    num
+                ) < max_value:
+                    # Add to excel file result #num
+                    num = int(num)
+                    file_name_result = os.path.join(
+                        path, self.data_filtered[num]["file"]
+                    )
                     if book is None:
                         file_name = Files.be_list_excel
                         book = xlsxwriter.Workbook(
                             file_name, {"nan_inf_to_errors": True}
                         )
                     excel = Excel(
-                        file_name=self.data_filtered[int(num)]["file"],
+                        file_name=file_name_result,
                         book=book,
+                        compare=self.compare,
                     )
                     excel.report()
-            else:
-                if num not in ("r", "q"):
-                    print(f"Invalid option {num}. Try again!")
+                    print(f"Added {file_name_result} to {Files.be_list_excel}")
+                case [num] if num.isdigit() and int(num) < max_value:
+                    # Report the result #num
+                    num = int(num)
+                    file_name_result = os.path.join(
+                        path, self.data_filtered[num]["file"]
+                    )
+                    rep = Report(file_name_result, compare=self.compare)
+                    rep.report()
+                case _:
+                    print("Invalid option. Try again!")
 
     def show_result(self, data: dict, title: str = "") -> None:
         def whites(n: int) -> str:

benchmark/Utils.py

@@ -146,3 +146,7 @@ class TextColor:
     ENDC = "\033[0m"
     BOLD = "\033[1m"
     UNDERLINE = "\033[4m"
+    WHITE = "\033[97m"
+    GREY = "\033[90m"
+    BLACK = "\033[90m"
+    DEFAULT = "\033[99m"

benchmark/scripts/be_benchmark.py

@@ -5,7 +5,7 @@ from benchmark.Arguments import Arguments
 
 
 def main(args_test=None):
-    arguments = Arguments()
+    arguments = Arguments(prog="be_benchmark")
     arguments.xset("score").xset("excel").xset("tex_output").xset("quiet")
     args = arguments.parse(args_test)
     benchmark = Benchmark(score=args.score, visualize=not args.quiet)

benchmark/scripts/be_build_best.py

@@ -21,5 +21,5 @@ def main(args_test=None):
         print(e)
     else:
         if args.report:
-            report = ReportBest(args.score, args.model, best=True, grid=False)
+            report = ReportBest(args.score, args.model, best=True)
             report.report()

benchmark/scripts/be_init_project.py

@@ -0,0 +1,33 @@
+#!/usr/bin/env python
+import os
+from benchmark.Utils import Files, Folders
+from benchmark.Arguments import Arguments
+
+
+def main(args_test=None):
+    arguments = Arguments(prog="be_init_project")
+    arguments.add_argument("project_name", help="Project name")
+    args = arguments.parse(args_test)
+    folders = []
+    folders.append(args.project_name)
+    folders.append(os.path.join(args.project_name, Folders.results))
+    folders.append(os.path.join(args.project_name, Folders.hidden_results))
+    folders.append(os.path.join(args.project_name, Folders.exreport))
+    folders.append(os.path.join(args.project_name, Folders.report))
+    folders.append(os.path.join(args.project_name, Folders.img))
+    try:
+        for folder in folders:
+            print(f"Creating folder {folder}")
+            os.makedirs(folder)
+    except FileExistsError as e:
+        print(e)
+        exit(1)
+    env_src = os.path.join(Folders.src(), "..", f"{Files.dot_env}.dist")
+    env_to = os.path.join(args.project_name, Files.dot_env)
+    os.system(f"cp {env_src} {env_to}")
+    print("Done!")
+    print(
+        "Please, edit .env file with your settings and add a datasets folder"
+    )
+    print("with an all.txt file with the datasets you want to use.")
+    print("In that folder you have to include all the datasets you'll use.")

benchmark/scripts/be_list.py

@@ -1,7 +1,6 @@
 #! /usr/bin/env python
-import os
 from benchmark.Results import Summary
-from benchmark.Utils import Folders, Files
+from benchmark.Utils import Files
 from benchmark.Arguments import Arguments
 
 """List experiments of a model
@@ -9,12 +8,12 @@ from benchmark.Arguments import Arguments
 
 
 def main(args_test=None):
-    arguments = Arguments()
+    arguments = Arguments(prog="be_list")
     arguments.xset("number").xset("model", required=False).xset("key")
-    arguments.xset("hidden").xset("nan").xset("score", required=False)
-    arguments.xset("excel")
+    arguments.xset("score", required=False).xset("compare").xset("hidden")
+    arguments.xset("nan")
     args = arguments.parse(args_test)
-    data = Summary(hidden=args.hidden)
+    data = Summary(hidden=args.hidden, compare=args.compare)
     data.acquire()
     try:
         data.list_results(
@@ -22,40 +21,12 @@ def main(args_test=None):
             model=args.model,
             sort_key=args.key,
             number=args.number,
+            nan=args.nan,
         )
-        is_test = args_test is not None
-        if not args.nan:
-            excel_generated = data.manage_results(args.excel, is_test)
-            if args.excel and excel_generated:
-                print(f"Generated file: {Files.be_list_excel}")
-                Files.open(Files.be_list_excel, is_test)
     except ValueError as e:
         print(e)
         return
-    if args.nan:
-        results_nan = []
-        results = data.get_results_criteria(
-            score=args.score,
-            model=args.model,
-            input_data=None,
-            sort_key=args.key,
-            number=args.number,
-        )
-        for result in results:
-            if result["metric"] != result["metric"]:
-                results_nan.append(result)
-        if results_nan != []:
-            print(
-                "\n"
-                + "*" * 30
-                + " Results with nan moved to hidden "
-                + "*" * 30
-            )
-            data.data_filtered = []
-            data.list_results(input_data=results_nan)
-            for result in results_nan:
-                name = result["file"]
-                os.rename(
-                    os.path.join(Folders.results, name),
-                    os.path.join(Folders.hidden_results, name),
-                )
+    excel_generated = data.manage_results()
+    if excel_generated:
+        print(f"Generated file: {Files.be_list_excel}")
+        Files.open(Files.be_list_excel, test=args_test is not None)

benchmark/scripts/be_main.py

@@ -10,15 +10,20 @@ from benchmark.Arguments import Arguments
 
 
 def main(args_test=None):
-    arguments = Arguments()
+    arguments = Arguments(prog="be_main")
     arguments.xset("stratified").xset("score").xset("model", mandatory=True)
     arguments.xset("n_folds").xset("platform").xset("quiet").xset("title")
-    arguments.xset("hyperparameters").xset("paramfile").xset("report")
-    arguments.xset("grid_paramfile").xset("dataset")
+    arguments.xset("report")
+    arguments.add_exclusive(
+        ["grid_paramfile", "best_paramfile", "hyperparameters"]
+    )
+    arguments.xset(
+        "dataset", overrides="title", const="Test with only one dataset"
+    )
     args = arguments.parse(args_test)
     report = args.report or args.dataset is not None
     if args.grid_paramfile:
-        args.paramfile = False
+        args.best_paramfile = False
     try:
         job = Experiment(
             score_name=args.score,
@@ -26,7 +31,7 @@ def main(args_test=None):
             stratified=args.stratified,
             datasets=Datasets(dataset_name=args.dataset),
             hyperparams_dict=args.hyperparameters,
-            hyperparams_file=args.paramfile,
+            hyperparams_file=args.best_paramfile,
             grid_paramfile=args.grid_paramfile,
             progress_bar=not args.quiet,
             platform=args.platform,

benchmark/scripts/be_report.py

@@ -5,46 +5,79 @@ from benchmark.Arguments import Arguments
 
 
 """Build report on screen of a result file, optionally generate excel and sql
-file, and can compare results of report with best results obtained by model
+file, and can compare results of report wibth best results obtained by model
 If no argument is set, displays the datasets and its characteristics
 """
 
 
 def main(args_test=None):
-    arguments = Arguments()
-    arguments.xset("file").xset("excel").xset("sql").xset("compare")
-    arguments.xset("best").xset("grid").xset("model", required=False)
-    arguments.xset("score", required=False)
+    is_test = args_test is not None
+    arguments = Arguments(prog="be_report")
+    arguments.add_subparser()
+    arguments.add_subparsers_options(
+        (
+            "best",
+            "Report best results obtained by any model/score. "
+            "See be_build_best",
+        ),
+        [
+            ("model", dict(required=False)),
+            ("score", dict(required=False)),
+        ],
+    )
+    arguments.add_subparsers_options(
+        (
+            "grid",
+            "Report grid results obtained by any model/score. "
+            "See be_build_grid",
+        ),
+        [
+            ("model", dict(required=False)),
+            ("score", dict(required=False)),
+        ],
+    )
+    arguments.add_subparsers_options(
+        ("file", "Report file results"),
+        [
+            ("file_name", {}),
+            ("excel", {}),
+            ("sql", {}),
+            ("compare", {}),
+        ],
+    )
+    arguments.add_subparsers_options(
+        ("datasets", "Report datasets information"),
+        [
+            ("excel", {}),
+        ],
+    )
     args = arguments.parse(args_test)
-    if args.best:
-        args.grid = None
-    if args.grid:
-        args.best = None
-    if args.file is None and args.best is None and args.grid is None:
-        report = ReportDatasets(args.excel)
-        report.report()
-        if args.excel:
-            is_test = args_test is not None
-            Files.open(report.get_file_name(), is_test)
-    else:
-        if args.best is not None or args.grid is not None:
-            report = ReportBest(args.score, args.model, args.best, args.grid)
+    match args.subcommand:
+        case "best" | "grid":
+            best = args.subcommand == "best"
+            report = ReportBest(args.score, args.model, best)
             report.report()
-        else:
+        case "file":
             try:
-                report = Report(args.file, args.compare)
+                report = Report(args.file_name, args.compare)
+                report.report()
             except FileNotFoundError as e:
                 print(e)
-            else:
-                report.report()
-                if args.excel:
-                    excel = Excel(
-                        file_name=args.file,
-                        compare=args.compare,
-                    )
-                    excel.report()
-                    is_test = args_test is not None
-                    Files.open(excel.get_file_name(), is_test)
-                if args.sql:
-                    sql = SQL(args.file)
-                    sql.report()
+                return
+            if args.sql:
+                sql = SQL(args.file_name)
+                sql.report()
+            if args.excel:
+                excel = Excel(
+                    file_name=args.file_name,
+                    compare=args.compare,
+                )
+                excel.report()
+                Files.open(excel.get_file_name(), is_test)
+        case "datasets":
+            report = ReportDatasets(args.excel)
+            report.report()
+            if args.excel:
+                Files.open(report.get_file_name(), is_test)
+        case _:
+            arguments.print_help()

benchmark/tests/Arguments_test.py

@@ -24,13 +24,11 @@ class ArgumentsTest(TestBase):
 
     def test_parameters(self):
         expected_parameters = {
-            "best": ("-b", "--best"),
+            "best_paramfile": ("-b", "--best_paramfile"),
             "color": ("-c", "--color"),
             "compare": ("-c", "--compare"),
             "dataset": ("-d", "--dataset"),
             "excel": ("-x", "--excel"),
-            "file": ("-f", "--file"),
-            "grid": ("-g", "--grid"),
             "grid_paramfile": ("-g", "--grid_paramfile"),
             "hidden": ("--hidden",),
             "hyperparameters": ("-p", "--hyperparameters"),
@@ -42,7 +40,6 @@ class ArgumentsTest(TestBase):
             "nan": ("--nan",),
             "number": ("-n", "--number"),
             "n_folds": ("-n", "--n_folds"),
-            "paramfile": ("-f", "--paramfile"),
             "platform": ("-P", "--platform"),
             "quiet": ("-q", "--quiet"),
             "report": ("-r", "--report"),
@@ -98,3 +95,27 @@ class ArgumentsTest(TestBase):
         finally:
             os.chdir(path)
         self.assertEqual(stderr.getvalue(), f"{NO_ENV}\n")
+
+    @patch("sys.stderr", new_callable=StringIO)
+    def test_overrides(self, stderr):
+        arguments = self.build_args()
+        arguments.xset("title")
+        arguments.xset("dataset", overrides="title", const="sample text")
+        test_args = ["-n", "3", "-m", "SVC", "-k", "1", "-d", "dataset"]
+        args = arguments.parse(test_args)
+        self.assertEqual(stderr.getvalue(), "")
+        self.assertEqual(args.title, "sample text")
+
+    @patch("sys.stderr", new_callable=StringIO)
+    def test_overrides_no_args(self, stderr):
+        arguments = self.build_args()
+        arguments.xset("title")
+        arguments.xset("dataset", overrides="title", const="sample text")
+        test_args = None
+        with self.assertRaises(SystemExit):
+            arguments.parse(test_args)
+        self.assertRegexpMatches(
+            stderr.getvalue(),
+            r"error: the following arguments are required: -m/--model, "
+            "-k/--key, --title",
+        )

benchmark/tests/Dataset_test.py

@@ -33,8 +33,8 @@ class DatasetTest(TestBase):
     def test_load_dataframe(self):
         self.set_env(".env.arff")
         dt = Datasets()
-        X, y = dt.load_discretized("iris", dataframe=False)
-        dataset = dt.load_discretized("iris", dataframe=True)
+        X, y = dt.load("iris", dataframe=False)
+        dataset = dt.load("iris", dataframe=True)
         class_name = dt.get_class_name()
         features = dt.get_features()
         self.assertListEqual(y.tolist(), dataset[class_name].tolist())

benchmark/tests/Report_test.py

@@ -69,13 +69,13 @@ class ReportTest(TestBase):
             _ = Report("unknown_file")
 
     def test_report_best(self):
-        report = ReportBest("accuracy", "STree", best=True, grid=False)
+        report = ReportBest("accuracy", "STree", best=True)
         with patch(self.output, new=StringIO()) as stdout:
             report.report()
         self.check_output_file(stdout, "report_best")
 
     def test_report_grid(self):
-        report = ReportBest("accuracy", "STree", best=False, grid=True)
+        report = ReportBest("accuracy", "STree", best=False)
         with patch(self.output, new=StringIO()) as stdout:
             report.report()
         file_name = "report_grid.test"
@@ -90,12 +90,6 @@ class ReportTest(TestBase):
 
             self.assertEqual(line, output_text[index])
 
-    def test_report_best_both(self):
-        report = ReportBest("accuracy", "STree", best=True, grid=True)
-        with patch(self.output, new=StringIO()) as stdout:
-            report.report()
-        self.check_output_file(stdout, "report_best")
-
     @patch("sys.stdout", new_callable=StringIO)
     def test_report_datasets(self, mock_output):
         report = ReportDatasets()

benchmark/tests/__init__.py

@@ -13,6 +13,7 @@ from .PairCheck_test import PairCheckTest
 from .Arguments_test import ArgumentsTest
 from .scripts.Be_Pair_check_test import BePairCheckTest
 from .scripts.Be_List_test import BeListTest
+from .scripts.Be_Init_Project_test import BeInitProjectTest
 from .scripts.Be_Report_test import BeReportTest
 from .scripts.Be_Summary_test import BeSummaryTest
 from .scripts.Be_Grid_test import BeGridTest

benchmark/tests/scripts/Be_Benchmark_test.py

@@ -25,7 +25,7 @@ class BeBenchmarkTest(TestBase):
 
     def test_be_benchmark_complete(self):
         stdout, stderr = self.execute_script(
-            "be_benchmark", ["-s", self.score, "-q", "1", "-t", "1", "-x", "1"]
+            "be_benchmark", ["-s", self.score, "-q", "-t", "-x"]
         )
         self.assertEqual(stderr.getvalue(), "")
         # Check output
@@ -60,7 +60,7 @@ class BeBenchmarkTest(TestBase):
 
     def test_be_benchmark_single(self):
         stdout, stderr = self.execute_script(
-            "be_benchmark", ["-s", self.score, "-q", "1"]
+            "be_benchmark", ["-s", self.score, "-q"]
         )
         self.assertEqual(stderr.getvalue(), "")
         # Check output

benchmark/tests/scripts/Be_Best_test.py

@@ -67,7 +67,7 @@ class BeBestTest(TestBase):
 
     def test_be_build_best_report(self):
         stdout, _ = self.execute_script(
-            "be_build_best", ["-s", "accuracy", "-m", "ODTE", "-r", "1"]
+            "be_build_best", ["-s", "accuracy", "-m", "ODTE", "-r"]
         )
         expected_data = {
             "balance-scale": [

benchmark/tests/scripts/Be_Grid_test.py

@@ -69,7 +69,7 @@ class BeGridTest(TestBase):
     def test_be_grid_no_input(self):
         stdout, stderr = self.execute_script(
             "be_grid",
-            ["-m", "ODTE", "-s", "f1-weighted", "-q", "1"],
+            ["-m", "ODTE", "-s", "f1-weighted", "-q"],
         )
         self.assertEqual(stderr.getvalue(), "")
         grid_file = os.path.join(

benchmark/tests/scripts/Be_Init_Project_test.py

@@ -0,0 +1,66 @@
+import os
+from io import StringIO
+from unittest.mock import patch
+from ..TestBase import TestBase
+from ...Utils import Folders
+
+
+class BeInitProjectTest(TestBase):
+    def setUp(self):
+        self.prepare_scripts_env()
+
+    def tearDown(self):
+        if os.path.exists("test_project"):
+            os.system("rm -rf test_project")
+
+    def assertIsFile(self, file_name):
+        if not os.path.isfile(file_name):
+            raise AssertionError(f"File {str(file_name)} does not exist")
+
+    def assertIsFolder(self, path):
+        if not os.path.exists(path):
+            raise AssertionError(f"Folder {str(path)} does not exist")
+
+    def test_be_init_project(self):
+        test_project = "test_project"
+        stdout, stderr = self.execute_script("be_init_project", [test_project])
+        self.assertEqual(stderr.getvalue(), "")
+        self.check_output_file(stdout, "be_init_project")
+        # check folders
+        expected = [
+            Folders.results,
+            Folders.hidden_results,
+            Folders.exreport,
+            Folders.report,
+            Folders.img,
+        ]
+        for folder in expected:
+            self.assertIsFolder(os.path.join(test_project, folder))
+        self.assertIsFile(os.path.join(test_project, ".env"))
+        os.system(f"rm -rf {test_project}")
+
+    @patch("sys.stdout", new_callable=StringIO)
+    @patch("sys.stderr", new_callable=StringIO)
+    def test_be_init_project_no_arguments(self, stdout, stderr):
+        with self.assertRaises(SystemExit) as cm:
+            module = self.search_script("be_init_project")
+            module.main("")
+        self.assertEqual(cm.exception.code, 2)
+        self.check_output_file(stdout, "be_init_project_no_arguments")
+        self.assertEqual(stderr.getvalue(), "")
+
+    @patch("sys.stdout", new_callable=StringIO)
+    @patch("sys.stderr", new_callable=StringIO)
+    def test_be_init_project_twice(self, stdout, stderr):
+        test_project = "test_project"
+        self.execute_script("be_init_project", [test_project])
+        with self.assertRaises(SystemExit) as cm:
+            module = self.search_script("be_init_project")
+            module.main([test_project])
+        self.assertEqual(cm.exception.code, 1)
+        self.assertEqual(
+            stderr.getvalue(),
+            f"Creating folder {test_project}\n"
+            f"[Errno 17] File exists: '{test_project}'\n",
+        )
+        self.assertEqual(stdout.getvalue(), "")

benchmark/tests/scripts/Be_List_test.py

@@ -1,4 +1,5 @@
 import os
+import shutil
 from unittest.mock import patch
 from openpyxl import load_workbook
 from ...Utils import Folders, Files, NO_RESULTS
@@ -27,36 +28,26 @@ class BeListTest(TestBase):
         self.assertEqual(stderr.getvalue(), "")
         self.check_output_file(stdout, "be_list_report")
 
-    @patch("benchmark.Results.get_input", side_effect=iter(["q"]))
-    def test_be_list_report_excel_none(self, input_data):
-        stdout, stderr = self.execute_script(
-            "be_list", ["-m", "STree", "-x", "1"]
-        )
-        self.assertEqual(stderr.getvalue(), "")
-        self.check_output_file(stdout, "be_list_model")
-
     @patch("benchmark.Results.get_input", side_effect=iter(["r", "q"]))
     def test_be_list_twice(self, input_data):
         stdout, stderr = self.execute_script("be_list", ["-m", "STree"])
         self.assertEqual(stderr.getvalue(), "")
         self.check_output_file(stdout, "be_list_model_2")
 
-    @patch("benchmark.Results.get_input", side_effect=iter(["2", "q"]))
+    @patch("benchmark.Results.get_input", side_effect=iter(["e 2", "q"]))
     def test_be_list_report_excel(self, input_data):
-        stdout, stderr = self.execute_script(
-            "be_list", ["-m", "STree", "-x", "1"]
-        )
+        stdout, stderr = self.execute_script("be_list", ["-m", "STree"])
         self.assertEqual(stderr.getvalue(), "")
         self.check_output_file(stdout, "be_list_report_excel")
         book = load_workbook(Files.be_list_excel)
         sheet = book["STree"]
         self.check_excel_sheet(sheet, "excel")
 
-    @patch("benchmark.Results.get_input", side_effect=iter(["2", "1", "q"]))
+    @patch(
+        "benchmark.Results.get_input", side_effect=iter(["e 2", "e 1", "q"])
+    )
     def test_be_list_report_excel_twice(self, input_data):
-        stdout, stderr = self.execute_script(
-            "be_list", ["-m", "STree", "-x", "1"]
-        )
+        stdout, stderr = self.execute_script("be_list", ["-m", "STree"])
         self.assertEqual(stderr.getvalue(), "")
         self.check_output_file(stdout, "be_list_report_excel_2")
         book = load_workbook(Files.be_list_excel)
@@ -73,7 +64,36 @@ class BeListTest(TestBase):
         self.assertEqual(stderr.getvalue(), "")
         self.assertEqual(stdout.getvalue(), f"{NO_RESULTS}\n")
 
-    def test_be_list_nan(self):
+    @patch(
+        "benchmark.Results.get_input", side_effect=iter(["d 0", "y", "", "q"])
+    )
+    # @patch("benchmark.Results.get_input", side_effect=iter(["q"]))
+    def test_be_list_delete(self, input_data):
+        def copy_files(source_folder, target_folder, file_name):
+            source = os.path.join(source_folder, file_name)
+            target = os.path.join(target_folder, file_name)
+            shutil.copyfile(source, target)
+
+        file_name = (
+            "results_accuracy_XGBoost_MacBookpro16_2022-05-04_11:00:"
+            "35_0.json"
+        )
+        # move nan result from hidden to results
+        copy_files(Folders.hidden_results, Folders.results, file_name)
+        try:
+            # list and delete result
+            stdout, stderr = self.execute_script("be_list", "")
+            self.assertEqual(stderr.getvalue(), "")
+            self.check_output_file(stdout, "be_list_delete")
+        except Exception:
+            # delete the result copied if be_list couldn't
+            os.unlink(os.path.join(Folders.results, file_name))
+            self.fail("test_be_list_delete() should not raise exception")
+
+    @patch(
+        "benchmark.Results.get_input", side_effect=iter(["h 0", "y", "", "q"])
+    )
+    def test_be_list_hide(self, input_data):
         def swap_files(source_folder, target_folder, file_name):
             source = os.path.join(source_folder, file_name)
             target = os.path.join(target_folder, file_name)
@@ -86,20 +106,38 @@ class BeListTest(TestBase):
         # move nan result from hidden to results
         swap_files(Folders.hidden_results, Folders.results, file_name)
         try:
-            # list and move nan result to hidden
-            stdout, stderr = self.execute_script("be_list", ["--nan", "1"])
+            # list and move nan result to hidden again
+            stdout, stderr = self.execute_script("be_list", "")
             self.assertEqual(stderr.getvalue(), "")
-            self.check_output_file(stdout, "be_list_nan")
+            self.check_output_file(stdout, "be_list_hide")
         except Exception:
-            # move back nan result file if be_list couldn't
+            # delete the result copied if be_list couldn't
             swap_files(Folders.results, Folders.hidden_results, file_name)
-            self.fail("test_be_list_nan() should not raise exception")
+            self.fail("test_be_list_hide() should not raise exception")
 
-    @patch("benchmark.Results.get_input", return_value="q")
-    def test_be_list_nan_no_nan(self, input_data):
-        stdout, stderr = self.execute_script("be_list", ["--nan", "1"])
+    @patch("benchmark.Results.get_input", side_effect=iter(["h 0", "q"]))
+    def test_be_list_already_hidden(self, input_data):
+        stdout, stderr = self.execute_script("be_list", ["--hidden"])
         self.assertEqual(stderr.getvalue(), "")
-        self.check_output_file(stdout, "be_list_no_nan")
+        self.check_output_file(stdout, "be_list_already_hidden")
+
+    @patch("benchmark.Results.get_input", side_effect=iter(["h 0", "n", "q"]))
+    def test_be_list_dont_hide(self, input_data):
+        stdout, stderr = self.execute_script("be_list", "")
+        self.assertEqual(stderr.getvalue(), "")
+        self.check_output_file(stdout, "be_list_default")
+
+    @patch("benchmark.Results.get_input", side_effect=iter(["q"]))
+    def test_be_list_hidden_nan(self, input_data):
+        stdout, stderr = self.execute_script("be_list", ["--hidden", "--nan"])
+        self.assertEqual(stderr.getvalue(), "")
+        self.check_output_file(stdout, "be_list_hidden_nan")
+
+    @patch("benchmark.Results.get_input", side_effect=iter(["q"]))
+    def test_be_list_hidden(self, input_data):
+        stdout, stderr = self.execute_script("be_list", ["--hidden"])
+        self.assertEqual(stderr.getvalue(), "")
+        self.check_output_file(stdout, "be_list_hidden")
 
     def test_be_no_env(self):
         path = os.getcwd()

benchmark/tests/scripts/Be_Main_test.py

@@ -1,4 +1,5 @@
 import os
+import json
 from io import StringIO
 from unittest.mock import patch
 from ...Results import Report
@@ -30,7 +31,7 @@ class BeMainTest(TestBase):
     def test_be_main_complete(self):
         stdout, _ = self.execute_script(
             "be_main",
-            ["-s", self.score, "-m", "STree", "--title", "test", "-r", "1"],
+            ["-s", self.score, "-m", "STree", "--title", "test", "-r"],
         )
         # keep the report name to delete it after
         report_name = stdout.getvalue().splitlines()[-1].split("in ")[1]
@@ -66,10 +67,8 @@ class BeMainTest(TestBase):
                 "STree",
                 "--title",
                 "test",
-                "-f",
-                "1",
+                "-b",
                 "-r",
-                "1",
             ],
         )
         # keep the report name to delete it after
@@ -79,6 +78,48 @@ class BeMainTest(TestBase):
             stdout, "be_main_best", [0, 2, 3, 5, 6, 7, 8, 9, 12, 13, 14]
         )
 
+    @patch("sys.stdout", new_callable=StringIO)
+    @patch("sys.stderr", new_callable=StringIO)
+    def test_be_main_incompatible_params(self, stdout, stderr):
+        m1 = (
+            "be_main: error: argument -b/--best_paramfile: not allowed with "
+            "argument -p/--hyperparameters"
+        )
+        m2 = (
+            "be_main: error: argument -g/--grid_paramfile: not allowed with "
+            "argument -p/--hyperparameters"
+        )
+        m3 = (
+            "be_main: error: argument -g/--grid_paramfile: not allowed with "
+            "argument -p/--hyperparameters"
+        )
+        m4 = m1
+        p0 = [
+            "-s",
+            self.score,
+            "-m",
+            "SVC",
+            "--title",
+            "test",
+        ]
+        pset = json.dumps(dict(C=17))
+        p1 = p0.copy()
+        p1.extend(["-p", pset, "-b"])
+        p2 = p0.copy()
+        p2.extend(["-p", pset, "-g"])
+        p3 = p0.copy()
+        p3.extend(["-p", pset, "-g", "-b"])
+        p4 = p0.copy()
+        p4.extend(["-b", "-g"])
+        parameters = [(p1, m1), (p2, m2), (p3, m3), (p4, m4)]
+        for parameter, message in parameters:
+            with self.assertRaises(SystemExit) as msg:
+                module = self.search_script("be_main")
+                module.main(parameter)
+            self.assertEqual(msg.exception.code, 2)
+            self.assertEqual(stderr.getvalue(), "")
+            self.assertRegexpMatches(stdout.getvalue(), message)
+
     def test_be_main_best_params_non_existent(self):
         model = "GBC"
         stdout, stderr = self.execute_script(
@@ -90,10 +131,8 @@ class BeMainTest(TestBase):
                 model,
                 "--title",
                 "test",
-                "-f",
-                "1",
+                "-b",
                 "-r",
-                "1",
             ],
         )
         self.assertEqual(stderr.getvalue(), "")
@@ -117,9 +156,7 @@ class BeMainTest(TestBase):
                 "--title",
                 "test",
                 "-g",
-                "1",
                 "-r",
-                "1",
             ],
         )
         self.assertEqual(stderr.getvalue(), "")
@@ -142,9 +179,7 @@ class BeMainTest(TestBase):
                 "--title",
                 "test",
                 "-g",
-                "1",
                 "-r",
-                "1",
             ],
         )
         # keep the report name to delete it after

benchmark/tests/scripts/Be_Print_Strees_test.py

@@ -18,7 +18,7 @@ class BePrintStrees(TestBase):
         for name in self.datasets:
             stdout, _ = self.execute_script(
                 "be_print_strees",
-                ["-d", name, "-q", "1"],
+                ["-d", name, "-q"],
             )
             file_name = os.path.join(Folders.img, f"stree_{name}.png")
             self.files.append(file_name)
@@ -33,7 +33,7 @@ class BePrintStrees(TestBase):
         for name in self.datasets:
             stdout, _ = self.execute_script(
                 "be_print_strees",
-                ["-d", name, "-q", "1", "-c", "1"],
+                ["-d", name, "-q", "-c"],
             )
             file_name = os.path.join(Folders.img, f"stree_{name}.png")
             self.files.append(file_name)

benchmark/tests/scripts/Be_Report_test.py

@@ -1,5 +1,7 @@
 import os
 from openpyxl import load_workbook
+from io import StringIO
+from unittest.mock import patch
 from ...Utils import Folders, Files
 from ..TestBase import TestBase
 from ..._version import __version__
@@ -23,25 +25,25 @@ class BeReportTest(TestBase):
             "results",
             "results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json",
         )
-        stdout, stderr = self.execute_script("be_report", ["-f", file_name])
+        stdout, stderr = self.execute_script("be_report", ["file", file_name])
         self.assertEqual(stderr.getvalue(), "")
         self.check_output_file(stdout, "report")
 
     def test_be_report_not_found(self):
-        stdout, stderr = self.execute_script("be_report", ["-f", "unknown"])
+        stdout, stderr = self.execute_script("be_report", ["file", "unknown"])
         self.assertEqual(stderr.getvalue(), "")
         self.assertEqual(stdout.getvalue(), "unknown does not exists!\n")
 
     def test_be_report_compare(self):
         file_name = "results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json"
         stdout, stderr = self.execute_script(
-            "be_report", ["-f", file_name, "-c", "1"]
+            "be_report", ["file", file_name, "-c"]
         )
         self.assertEqual(stderr.getvalue(), "")
         self.check_output_file(stdout, "report_compared")
 
     def test_be_report_datatsets(self):
-        stdout, stderr = self.execute_script("be_report", [])
+        stdout, stderr = self.execute_script("be_report", ["datasets"])
         self.assertEqual(stderr.getvalue(), "")
         file_name = f"report_datasets{self.ext}"
         with open(os.path.join(self.test_files, file_name)) as f:
@@ -54,7 +56,7 @@ class BeReportTest(TestBase):
             self.assertEqual(line, output_text[index])
 
     def test_be_report_datasets_excel(self):
-        stdout, stderr = self.execute_script("be_report", ["-x", "1"])
+        stdout, stderr = self.execute_script("be_report", ["datasets", "-x"])
         self.assertEqual(stderr.getvalue(), "")
         file_name = f"report_datasets{self.ext}"
         with open(os.path.join(self.test_files, file_name)) as f:
@@ -77,14 +79,14 @@ class BeReportTest(TestBase):
 
     def test_be_report_best(self):
         stdout, stderr = self.execute_script(
-            "be_report", ["-s", "accuracy", "-m", "STree", "-b", "1"]
+            "be_report", ["best", "-s", "accuracy", "-m", "STree"]
         )
         self.assertEqual(stderr.getvalue(), "")
         self.check_output_file(stdout, "report_best")
 
     def test_be_report_grid(self):
         stdout, stderr = self.execute_script(
-            "be_report", ["-s", "accuracy", "-m", "STree", "-g", "1"]
+            "be_report", ["grid", "-s", "accuracy", "-m", "STree"]
         )
         self.assertEqual(stderr.getvalue(), "")
         file_name = "report_grid.test"
@@ -98,19 +100,24 @@ class BeReportTest(TestBase):
                 line = self.replace_STree_version(line, output_text, index)
             self.assertEqual(line, output_text[index])
 
-    def test_be_report_best_both(self):
-        stdout, stderr = self.execute_script(
-            "be_report",
-            ["-s", "accuracy", "-m", "STree", "-b", "1", "-g", "1"],
-        )
+    @patch("sys.stderr", new_callable=StringIO)
+    def test_be_report_unknown_subcommand(self, stderr):
+        with self.assertRaises(SystemExit) as msg:
+            module = self.search_script("be_report")
+            module.main(["unknown"])
+        self.assertEqual(msg.exception.code, 2)
+        self.check_output_file(stderr, "report_unknown_subcommand")
+
+    def test_be_report_without_subcommand(self):
+        stdout, stderr = self.execute_script("be_report", "")
         self.assertEqual(stderr.getvalue(), "")
-        self.check_output_file(stdout, "report_best")
+        self.check_output_file(stdout, "report_without_subcommand")
 
     def test_be_report_excel_compared(self):
         file_name = "results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json"
         stdout, stderr = self.execute_script(
             "be_report",
-            ["-f", file_name, "-x", "1", "-c", "1"],
+            ["file", file_name, "-x", "-c"],
         )
         file_name = os.path.join(
             Folders.results, file_name.replace(".json", ".xlsx")
@@ -125,7 +132,7 @@ class BeReportTest(TestBase):
         file_name = "results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json"
         stdout, stderr = self.execute_script(
             "be_report",
-            ["-f", file_name, "-x", "1"],
+            ["file", file_name, "-x"],
         )
         file_name = os.path.join(
             Folders.results, file_name.replace(".json", ".xlsx")
@@ -140,7 +147,7 @@ class BeReportTest(TestBase):
         file_name = "results_accuracy_ODTE_Galgo_2022-04-20_10:52:20_0.json"
         stdout, stderr = self.execute_script(
             "be_report",
-            ["-f", file_name, "-q", "1"],
+            ["file", file_name, "-q"],
         )
         file_name = os.path.join(
             Folders.results, file_name.replace(".json", ".sql")

benchmark/tests/test_files/be_init_project.test

@@ -0,0 +1,10 @@
+Creating folder test_project
+Creating folder test_project/results
+Creating folder test_project/hidden_results
+Creating folder test_project/exreport
+Creating folder test_project/exreport/exreport_output
+Creating folder test_project/img
+Done!
+Please, edit .env file with your settings and add a datasets folder
+with an all.txt file with the datasets you want to use.
+In that folder you have to include all the datasets you'll use.

benchmark/tests/test_files/be_init_project_no_arguments.test

@@ -0,0 +1,2 @@
+usage: be_init_project [-h] project_name
+be_init_project: error: the following arguments are required: project_name

benchmark/tests/test_files/be_list_already_hidden.test

@@ -0,0 +1,5 @@
+ #  Date       File                                                             Score    Time(h) Title
+=== ========== ================================================================ ======== ======= =======================
+  0 2022-05-04 results_accuracy_XGBoost_MacBookpro16_2022-05-04_11:00:35_0.json      nan   3.091 Default hyperparameters
+  1 2021-11-01 results_accuracy_STree_iMac27_2021-11-01_23:55:16_0.json          0.97446   0.098 default
+Already hidden

benchmark/tests/test_files/be_list_delete.test

@@ -0,0 +1,16 @@
+ #  Date       File                                                             Score    Time(h) Title
+=== ========== ================================================================ ======== ======= ============================================
+  0 2022-05-04 results_accuracy_XGBoost_MacBookpro16_2022-05-04_11:00:35_0.json      nan   3.091 Default hyperparameters
+  1 2022-04-20 results_accuracy_ODTE_Galgo_2022-04-20_10:52:20_0.json            0.04341   6.275 Gridsearched hyperparams v022.1b random_init
+  2 2022-01-14 results_accuracy_RandomForest_iMac27_2022-01-14_12:39:30_0.json   0.03627   0.076 Test default paramters with RandomForest
+  3 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json     0.03790   1.143 default B
+  4 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json          0.04158   0.943 default A
+  5 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json          0.04544   0.173 With gridsearched hyperparameters
+Deleting results/results_accuracy_XGBoost_MacBookpro16_2022-05-04_11:00:35_0.json
+ #  Date       File                                                            Score    Time(h) Title
+=== ========== =============================================================== ======== ======= ============================================
+  0 2022-04-20 results_accuracy_ODTE_Galgo_2022-04-20_10:52:20_0.json           0.04341   6.275 Gridsearched hyperparams v022.1b random_init
+  1 2022-01-14 results_accuracy_RandomForest_iMac27_2022-01-14_12:39:30_0.json  0.03627   0.076 Test default paramters with RandomForest
+  2 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json    0.03790   1.143 default B
+  3 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json         0.04158   0.943 default A
+  4 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json         0.04544   0.173 With gridsearched hyperparameters

benchmark/tests/test_files/be_list_hidden.test

@@ -0,0 +1,4 @@
+ #  Date       File                                                             Score    Time(h) Title
+=== ========== ================================================================ ======== ======= =======================
+  0 2022-05-04 results_accuracy_XGBoost_MacBookpro16_2022-05-04_11:00:35_0.json      nan   3.091 Default hyperparameters
+  1 2021-11-01 results_accuracy_STree_iMac27_2021-11-01_23:55:16_0.json          0.97446   0.098 default

benchmark/tests/test_files/be_list_hidden_nan.test

@@ -0,0 +1,3 @@
+ #  Date       File                                                             Score    Time(h) Title
+=== ========== ================================================================ ======== ======= =======================
+  0 2022-05-04 results_accuracy_XGBoost_MacBookpro16_2022-05-04_11:00:35_0.json      nan   3.091 Default hyperparameters

benchmark/tests/test_files/be_list_hide.test

@@ -6,8 +6,11 @@
   3 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json     0.03790   1.143 default B
   4 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json          0.04158   0.943 default A
   5 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json          0.04544   0.173 With gridsearched hyperparameters
-
-****************************** Results with nan moved to hidden ******************************
- #  Date       File                                                             Score    Time(h) Title
-=== ========== ================================================================ ======== ======= =======================
-  0 2022-05-04 results_accuracy_XGBoost_MacBookpro16_2022-05-04_11:00:35_0.json      nan   3.091 Default hyperparameters
+Hiding results/results_accuracy_XGBoost_MacBookpro16_2022-05-04_11:00:35_0.json
+ #  Date       File                                                            Score    Time(h) Title
+=== ========== =============================================================== ======== ======= ============================================
+  0 2022-04-20 results_accuracy_ODTE_Galgo_2022-04-20_10:52:20_0.json           0.04341   6.275 Gridsearched hyperparams v022.1b random_init
+  1 2022-01-14 results_accuracy_RandomForest_iMac27_2022-01-14_12:39:30_0.json  0.03627   0.076 Test default paramters with RandomForest
+  2 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json    0.03790   1.143 default B
+  3 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json         0.04158   0.943 default A
+  4 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json         0.04544   0.173 With gridsearched hyperparameters

benchmark/tests/test_files/be_list_model.test

@@ -3,4 +3,3 @@
   0 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json  0.03790   1.143 default B
   1 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json       0.04158   0.943 default A
   2 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json       0.04544   0.173 With gridsearched hyperparameters
-Which result do you want to report? (q to quit, r to list again, number to report): 

benchmark/tests/test_files/be_list_model_2.test

@@ -3,9 +3,8 @@
   0 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json  0.03790   1.143 default B
   1 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json       0.04158   0.943 default A
   2 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json       0.04544   0.173 With gridsearched hyperparameters
-Which result do you want to report? (q to quit, r to list again, number to report):  #  Date       File                                                          Score    Time(h) Title
+ #  Date       File                                                          Score    Time(h) Title
 === ========== ============================================================= ======== ======= =================================
   0 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json  0.03790   1.143 default B
   1 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json       0.04158   0.943 default A
   2 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json       0.04544   0.173 With gridsearched hyperparameters
-Which result do you want to report? (q to quit, r to list again, number to report): 

benchmark/tests/test_files/be_list_model_invalid.test

@@ -3,5 +3,4 @@
   0 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json  0.03790   1.143 default B
   1 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json       0.04158   0.943 default A
   2 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json       0.04544   0.173 With gridsearched hyperparameters
-Which result do you want to report? (q to quit, r to list again, number to report): Invalid option x. Try again!
-Which result do you want to report? (q to quit, r to list again, number to report): 
+Invalid option. Try again!

benchmark/tests/test_files/be_list_report.test

@@ -3,7 +3,7 @@
   0 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json  0.03790   1.143 default B
   1 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json       0.04158   0.943 default A
   2 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json       0.04544   0.173 With gridsearched hyperparameters
-Which result do you want to report? (q to quit, r to list again, number to report): *************************************************************************************************************************
+*************************************************************************************************************************
 * STree ver. 1.2.3 Python ver. 3.11x with 5 Folds cross validation and 10 random seeds. 2021-11-01 19:17:07             *
 * default B                                                                                                             *
 * Random seeds: [57, 31, 1714, 17, 23, 79, 83, 97, 7, 1] Stratified: False                                              *
@@ -18,4 +18,3 @@ Dataset                        Sampl. Feat. Cls Nodes   Leaves  Depth   Score
 *************************************************************************************************************************
 * accuracy compared to STree_default (liblinear-ovr) .:  0.0379                                                         *
 *************************************************************************************************************************
-Which result do you want to report? (q to quit, r to list again, number to report): 

benchmark/tests/test_files/be_list_report_excel.test

@@ -3,19 +3,5 @@
   0 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json  0.03790   1.143 default B
   1 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json       0.04158   0.943 default A
   2 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json       0.04544   0.173 With gridsearched hyperparameters
-Which result do you want to report? (q to quit, r to list again, number to report): *************************************************************************************************************************
-* STree ver. 1.2.3 Python ver. 3.11x with 5 Folds cross validation and 10 random seeds. 2021-09-30 11:42:07             *
-* With gridsearched hyperparameters                                                                                     *
-* Random seeds: [57, 31, 1714, 17, 23, 79, 83, 97, 7, 1] Stratified: False                                              *
-* Execution took  624.25 seconds,   0.17 hours, on iMac27                                                               *
-* Score is accuracy                                                                                                     *
-*************************************************************************************************************************
-
-Dataset                        Sampl. Feat. Cls Nodes   Leaves  Depth   Score           Time              Hyperparameters 
-============================== ====== ===== === ======= ======= ======= =============== ================= =============== 
-balance-scale                     625     4   3    7.00    4.00    3.00 0.970560±0.0150   0.014049±0.0020 {'C': 10000.0, 'gamma': 0.1, 'kernel': 'rbf', 'max_iter': 10000.0, 'multiclass_strategy': 'ovr'} 
-balloons                           16     4   2    3.00    2.00    2.00 0.860000±0.2850   0.000854±0.0000 {'C': 7, 'gamma': 0.1, 'kernel': 'rbf', 'max_iter': 10000.0, 'multiclass_strategy': 'ovr'} 
-*************************************************************************************************************************
-* accuracy compared to STree_default (liblinear-ovr) .:  0.0454                                                         *
-*************************************************************************************************************************
-Which result do you want to report? (q to quit, r to list again, number to report): Generated file: some_results.xlsx
+Added results/results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json to some_results.xlsx
+Generated file: some_results.xlsx

benchmark/tests/test_files/be_list_report_excel_2.test

@@ -3,34 +3,6 @@
   0 2021-11-01 results_accuracy_STree_macbook-pro_2021-11-01_19:17:07_0.json  0.03790   1.143 default B
   1 2021-10-27 results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json       0.04158   0.943 default A
   2 2021-09-30 results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json       0.04544   0.173 With gridsearched hyperparameters
-Which result do you want to report? (q to quit, r to list again, number to report): *************************************************************************************************************************
-* STree ver. 1.2.3 Python ver. 3.11x with 5 Folds cross validation and 10 random seeds. 2021-09-30 11:42:07             *
-* With gridsearched hyperparameters                                                                                     *
-* Random seeds: [57, 31, 1714, 17, 23, 79, 83, 97, 7, 1] Stratified: False                                              *
-* Execution took  624.25 seconds,   0.17 hours, on iMac27                                                               *
-* Score is accuracy                                                                                                     *
-*************************************************************************************************************************
-
-Dataset                        Sampl. Feat. Cls Nodes   Leaves  Depth   Score           Time              Hyperparameters 
-============================== ====== ===== === ======= ======= ======= =============== ================= =============== 
-balance-scale                     625     4   3    7.00    4.00    3.00 0.970560±0.0150   0.014049±0.0020 {'C': 10000.0, 'gamma': 0.1, 'kernel': 'rbf', 'max_iter': 10000.0, 'multiclass_strategy': 'ovr'} 
-balloons                           16     4   2    3.00    2.00    2.00 0.860000±0.2850   0.000854±0.0000 {'C': 7, 'gamma': 0.1, 'kernel': 'rbf', 'max_iter': 10000.0, 'multiclass_strategy': 'ovr'} 
-*************************************************************************************************************************
-* accuracy compared to STree_default (liblinear-ovr) .:  0.0454                                                         *
-*************************************************************************************************************************
-Which result do you want to report? (q to quit, r to list again, number to report): *************************************************************************************************************************
-* STree ver. 1.2.3 Python ver. 3.11x with 5 Folds cross validation and 10 random seeds. 2021-10-27 09:40:40             *
-* default A                                                                                                             *
-* Random seeds: [57, 31, 1714, 17, 23, 79, 83, 97, 7, 1] Stratified: False                                              *
-* Execution took 3395.01 seconds,   0.94 hours, on iMac27                                                               *
-* Score is accuracy                                                                                                     *
-*************************************************************************************************************************
-
-Dataset                        Sampl. Feat. Cls Nodes   Leaves  Depth   Score           Time              Hyperparameters 
-============================== ====== ===== === ======= ======= ======= =============== ================= =============== 
-balance-scale                     625     4   3   11.08    5.90    5.90 0.980000±0.0010   0.285207±0.0603 {'splitter': 'best', 'max_features': 'auto'} 
-balloons                           16     4   2    4.12    2.56    2.56 0.695000±0.2757   0.021201±0.0035 {'splitter': 'best', 'max_features': 'auto'} 
-*************************************************************************************************************************
-* accuracy compared to STree_default (liblinear-ovr) .:  0.0416                                                         *
-*************************************************************************************************************************
-Which result do you want to report? (q to quit, r to list again, number to report): Generated file: some_results.xlsx
+Added results/results_accuracy_STree_iMac27_2021-09-30_11:42:07_0.json to some_results.xlsx
+Added results/results_accuracy_STree_iMac27_2021-10-27_09:40:40_0.json to some_results.xlsx
+Generated file: some_results.xlsx

benchmark/tests/test_files/be_main_dataset.test

@@ -1,6 +1,6 @@
 *************************************************************************************************************************
 * STree ver. 1.2.4 Python ver. 3.11x with 5 Folds cross validation and 10 random seeds. 2022-05-08 19:38:28             *
-* test                                                                                                                  *
+* Test with only one dataset                                                                                            *
 * Random seeds: [57, 31, 1714, 17, 23, 79, 83, 97, 7, 1] Stratified: False                                              *
 * Execution took    0.06 seconds,   0.00 hours, on iMac27                                                               *
 * Score is accuracy                                                                                                     *

benchmark/tests/test_files/report_unknown_subcommand.test

@@ -0,0 +1,2 @@
+usage: be_report [-h] {best,grid,file,datasets} ...
+be_report: error: argument subcommand: invalid choice: 'unknown' (choose from 'best', 'grid', 'file', 'datasets')

benchmark/tests/test_files/report_without_subcommand.test

@@ -0,0 +1,12 @@
+usage: be_report [-h] {best,grid,file,datasets} ...
+
+positional arguments:
+  {best,grid,file,datasets}
+                        help for subcommand
+    best                Report best results obtained by any model/score. See be_build_best
+    grid                Report grid results obtained by any model/score. See be_build_grid
+    file                Report file results
+    datasets            Report datasets information
+
+options:
+  -h, --help            show this help message and exit

setup.py

@@ -38,6 +38,7 @@ def script_names():
         "print_strees",
         "report",
         "summary",
+        "init_project",
     ]
     result = []
     for script in scripts:
@@ -60,8 +61,6 @@ setuptools.setup(
     classifiers=[
         "Development Status :: 4 - Beta",
         "License :: OSI Approved :: " + get_data("license"),
-        "Programming Language :: Python :: 3.8",
-        "Programming Language :: Python :: 3.9",
         "Programming Language :: Python :: 3.10",
         "Natural Language :: English",
         "Topic :: Scientific/Engineering :: Artificial Intelligence",