Cannot import AutoML models from mlforecast.auto

[egolub22]

I am trying to create a solution that utilizes the library of "auto" models in mlforecast (https://nixtlaverse.nixtla.io/mlforecast/auto.html), in order to streamline hyper-parameter tuning. However, when I go to import the models in the way that is shown in Nixtla's tutorials (https://nixtlaverse.nixtla.io/mlforecast/docs/how-to-guides/hyperparameter_optimization.html) it indicates that the mlforecast.auto module does not exist. Indeed, when I checked what modules were in mlforecast, auto was not among them, and none of the existing modules contained any of the auto models.

`
Import statements:

import os
import numpy as np
import pandas as pd
import optuna
import lightgbm as lgb
import warnings
from importlib.metadata import version
from utilsforecast.evaluation import evaluate
from statsforecast import StatsForecast
from statsforecast.models import (
    AutoARIMA,
    AutoETS,
    AutoCES,
    AutoTheta,
    Theta,
    OptimizedTheta,
    DynamicTheta,
    DynamicOptimizedTheta,
    MSTL,
    SeasonalExponentialSmoothingOptimized)

from mlforecast import MLForecast
from mlforecast.target_transforms import Differences
from mlforecast.auto import (
    AutoLightGBM,
    AutoMLForecast,
    AutoModel,
    AutoRidge,
    ridge_space,
)

import plotly
from plotly.subplots import make_subplots
import plotly.graph_objs as go
from utilsforecast.losses import *

version('plotly')

`

Error:

`
Checking for packages:

import pkgutil

# this is the package we are inspecting -- for example 'email' from stdlib
import mlforecast

package = mlforecast
for importer, modname, ispkg in pkgutil.iter_modules(package.__path__):
    print("Found submodule %s (is a package: %s)" % (modname, ispkg))

`

Output:

[jmoralez]

The auto module was added in 0.12.1, which version are you using?

[egolub22]

I did not specify the version type when installing and it defaulted to an earlier version type for some reason. Thanks again for the help