Doctorado-ML/benchmark
1
0
Fork 0
mirror of https://github.com/Doctorado-ML/benchmark.git synced 2025-08-15 15:35:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix circular definition

Browse Source
This commit is contained in:
Ricardo Montañana Gómez
2022-05-04 02:07:41 +02:00
parent 90c895b616
commit e9d44e928b
7 changed files with 153 additions and 147 deletions
Download Patch File Download Diff File Expand all files Collapse all files

134
benchmark/Arguments.py
View File

@@ -11,137 +11,3 @@ ALL_METRICS = (
)
class EnvData:
@staticmethod
def load():
args = {}
with open(Files.dot_env) as f:
for line in f.read().splitlines():
if line == "" or line.startswith("#"):
continue
key, value = line.split("=")
args[key] = value
return args
class EnvDefault(argparse.Action):
# Thanks to https://stackoverflow.com/users/445507/russell-heilling
def __init__(self, envvar, required=True, default=None, **kwargs):
self._args = EnvData.load()
default = self._args[envvar]
required = False
super(EnvDefault, self).__init__(
default=default, required=required, **kwargs
)
def __call__(self, parser, namespace, values, option_string=None):
setattr(namespace, self.dest, values)
class Arguments:
def __init__(self):
self.ap = argparse.ArgumentParser()
models_data = Models.define_models(random_state=0)
models = "{" + ", ".join(models_data) + "}"
self.parameters = {
"best": [
("-b", "--best"),
{
"type": str,
"required": False,
"help": "best results of models",
},
],
"color": [],
"compare": [
("-c", "--compare"),
{
"type": bool,
"required": False,
"help": "Compare accuracy with best results",
},
],
"dataset": [],
"excel": [
("-x", "--excel"),
{
"type": bool,
"required": False,
"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": [],
"hidden": [],
"hyperparameters": [],
"key": [],
"lose": [],
"model": [
("-m", "--model"),
{
"type": str,
"required": True,
"choices": list(models_data),
"help": f"model name: {models}",
},
],
"model1": [],
"model2": [],
"nan": [],
"number": [],
"n_folds": [],
"paramfile": [],
"platform": [],
"quiet": [],
"report": [],
"score": [
("-s", "--score"),
{
"action": EnvDefault,
"envvar": "score",
"type": str,
"required": True,
"choices": ALL_METRICS,
},
],
"sql": [
("-q", "--sql"),
{"type": bool, "required": False, "help": "Generate SQL File"},
],
"stratified": [],
"tex_output": [
("-t", "--tex-output"),
{
"type": bool,
"required": False,
"default": False,
"help": "Generate Tex file with the table",
},
],
"title": [],
"win": [],
}
def xset(self, *arg_name, **kwargs):
print("parameters", arg_name[0])
names, default = self.parameters[arg_name[0]]
self.ap.add_argument(
*names,
**{**default, **kwargs},
)
return self
def parse(self):
return self.ap.parse_args()

3
benchmark/Experiments.py
View File

@@ -14,8 +14,7 @@ from sklearn.model_selection import (
cross_validate,
)
from .Utils import Folders, Files
from .Arguments import EnvData
from .Models import Models
from .Models import Models, EnvData
class Randomized:

146
benchmark/Models.py
View File

@@ -1,3 +1,4 @@
import argparse
from statistics import mean
from sklearn.tree import DecisionTreeClassifier, ExtraTreeClassifier
from sklearn.ensemble import (
@@ -11,6 +12,15 @@ from stree import Stree
from wodt import Wodt
from odte import Odte
from xgboost import XGBClassifier
from .Utils import Files
ALL_METRICS = (
"accuracy",
"f1-macro",
"f1-micro",
"f1-weighted",
"roc-auc-ovr",
)
class Models:
@@ -89,3 +99,139 @@ class Models:
nodes, leaves = result.nodes_leaves()
depth = result.depth_ if hasattr(result, "depth_") else 0
return nodes, leaves, depth
class EnvData:
@staticmethod
def load():
args = {}
with open(Files.dot_env) as f:
for line in f.read().splitlines():
if line == "" or line.startswith("#"):
continue
key, value = line.split("=")
args[key] = value
return args
class EnvDefault(argparse.Action):
# Thanks to https://stackoverflow.com/users/445507/russell-heilling
def __init__(self, envvar, required=True, default=None, **kwargs):
self._args = EnvData.load()
default = self._args[envvar]
required = False
super(EnvDefault, self).__init__(
default=default, required=required, **kwargs
)
def __call__(self, parser, namespace, values, option_string=None):
setattr(namespace, self.dest, values)
class Arguments:
def __init__(self):
self.ap = argparse.ArgumentParser()
models_data = Models.define_models(random_state=0)
self.parameters = {
"best": [
("-b", "--best"),
{
"type": str,
"required": False,
"help": "best results of models",
},
],
"color": [],
"compare": [
("-c", "--compare"),
{
"type": bool,
"required": False,
"help": "Compare accuracy with best results",
},
],
"dataset": [],
"excel": [
("-x", "--excel"),
{
"type": bool,
"required": False,
"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": [],
"hidden": [],
"hyperparameters": [],
"key": [],
"lose": [],
"model": [
("-m", "--model"),
{
"type": str,
"required": True,
"choices": list(models_data),
"action": EnvDefault,
"envvar": "model",
"help": f"model name",
},
],
"model1": [],
"model2": [],
"nan": [],
"number": [],
"n_folds": [],
"paramfile": [],
"platform": [],
"quiet": [],
"report": [],
"score": [
("-s", "--score"),
{
"action": EnvDefault,
"envvar": "score",
"type": str,
"required": True,
"choices": ALL_METRICS,
},
],
"sql": [
("-q", "--sql"),
{"type": bool, "required": False, "help": "Generate SQL File"},
],
"stratified": [],
"tex_output": [
("-t", "--tex-output"),
{
"type": bool,
"required": False,
"default": False,
"help": "Generate Tex file with the table",
},
],
"title": [],
"win": [],
}
def xset(self, *arg_name, **kwargs):
names, default = self.parameters[arg_name[0]]
self.ap.add_argument(
*names,
**{**default, **kwargs},
)
return self
def parse(self):
return self.ap.parse_args()

7
benchmark/Utils.py
View File

@@ -2,13 +2,6 @@ import os
import subprocess
BEST_ACCURACY_STREE = 40.282203
ALL_METRICS = (
"accuracy",
"f1-macro",
"f1-micro",
"f1-weighted",
"roc-auc-ovr",
)
class Folders:

2
benchmark/__init__.py
View File

@@ -1,6 +1,6 @@
from .Experiments import Experiment, Datasets, DatasetsSurcov, DatasetsTanveer
from .Results import Report, Summary
from .Arguments import EnvDefault
from .Models import EnvDefault
__author__ = "Ricardo Montañana Gómez"
__copyright__ = "Copyright 2020-2022, Ricardo Montañana Gómez"

2
benchmark/scripts/be_benchmark
View File

@@ -1,7 +1,7 @@
#!/usr/bin/env python
from benchmark.Results import Benchmark
from benchmark.Utils import Files
from benchmark.Arguments import Arguments
from benchmark.Models import Arguments
arguments = Arguments()

6
benchmark/scripts/be_report
View File

@@ -6,7 +6,7 @@ from benchmark.Utils import (
Files,
TextColor,
)
from benchmark.Arguments import Arguments
from benchmark.Models import Arguments
"""Build report on screen of a result file, optionally generate excel and sql
@@ -44,7 +44,9 @@ def default_report():
if __name__ == "__main__":
arguments = Arguments()
arguments.xset("file").xset("excel").xset("sql").xset("compare")
arguments.xset("best").xset("grid").xset("model").xset("score")
arguments.xset("best").xset("grid").xset("model", required=False).xset(
"score"
)
args = arguments.parse()
if args.grid: