def _print_models_overview(self):
print(f"\nIgel's supported models overview: \n")
reg_algs = list(models_dict.get('regression').keys())
clf_algs = list(models_dict.get('classification').keys())
cluster_algs = list(models_dict.get('clustering').keys())
df_algs = pd.DataFrame.from_dict({
"regression": reg_algs,
"classification": clf_algs,
"clustering": cluster_algs
}, orient="index").transpose().fillna('----')
df = self._tableize(df_algs)
print(df)
def _show_model_infos(self, model_name: str, model_type: str):
if not model_name:
print(f"Please enter a supported model")
self._print_models_overview()
else:
if not model_type:
print(
f"Please enter a type argument to get help on the chosen model\n"
f"type can be whether regression, classification or clustering \n"
)
self._print_models_overview()
return
if model_type not in ("regression", "classification",
"clustering"):
raise Exception(
f"{model_type} is not supported! \n"
f"model_type need to be regression, classification or clustering"
)
models = models_dict.get(model_type)
model_data = models.get(model_name)
model, link, *cv_class = model_data.values()
print(
f"model type: {model_type} \n"
f"model name: {model_name} \n"
f"sklearn model class: {model.__name__} \n"
f"{'-' * 60}\n"
f"You can click the link below to know more about the optional arguments\n"
f"that you can use with your chosen model ({model_name}).\n"
f"You can provide these optional arguments in the yaml file if you want to use them.\n"
f"link:\n{link} \n")
def models(self):
"""
show an overview of all models supported by igel
"""
if not self.dict_args or len(self.dict_args.keys()) <= 1:
self._print_models_overview()
else:
model_name = self.dict_args.get('model_name', None)
model_type = self.dict_args.get('model_type', None)
if not model_name:
print(f"Please enter a supported model")
self._print_models_overview()
else:
if not model_type:
print(
f"Please enter a type argument to get help on the chosen model\n"
f"type can be whether regression, classification or clustering \n"
)
self._print_models_overview()
return
if model_type not in ('regression', 'classification',
'clustering'):
raise Exception(
f"{model_type} is not supported! \n"
f"model_type need to be regression, classification or clustering"
)
models = models_dict.get(model_type)
model_data = models.get(model_name)
model, link, *cv_class = model_data.values()
print(
f"model type: {model_type} \n"
f"model name: {model_name} \n"
f"sklearn model class: {model.__name__} \n"
f"{'-' * 60}\n"
f"You can click the link below to know more about the optional arguments\n"
f"that you can use with your chosen model ({model_name}).\n"
f"You can provide these optional arguments in the yaml file if you want to use them.\n"
f"link:\n{Fore.GREEN}{link}{Fore.WHITE} \n")