安装

实测window安装不了，自己捣鼓了半天终于要成功时，报了个错

ImportError: No module named resource

pip安装也没解决，最后百度说是resource模块在windows上是不可用的
所以只能放弃

Mac 安装auto-pytorch

pip install swig
pip install pyrfr
pip install autoPyTorch

正常情况下就能安装好了

过程常见报错

1、lightgbm依赖

第一次使用报错：

OSError: dlopen(/Users/*/opt/anaconda3/lib/python3.9/site-packages/lightgbm/lib_lightgbm.so, 0x0006): Library not loaded: /usr/local/opt/libomp/lib/libomp.dylib
Referenced from: /Users//opt/anaconda3/lib/python3.9/site-packages/lightgbm/lib_lightgbm.so
Reason: tried: ‘/usr/local/opt/libomp/lib/libomp.dylib’ (no such file), ‘/usr/local/lib/libomp.dylib’ (no such file), ‘/usr/lib/libomp.dylib’ (no such file)

使用autoPyTorch需要安装lightgbm
解决方案：

pip uninstall lightgbm
conda install lightgbm

2、brew安装

bash: brew: command not found…

解决方案：
终端直接输入：

/bin/zsh -c "$(curl -fsSL https://gitee.com/cunkai/HomebrewCN/raw/master/Homebrew.sh)"

过程中，国内景象选择清华大学，差不多十几分钟就好了

3、brew install报错

fatal: not in a git directory Error: Command failed with exit 128: git

先输入：brew -v
会提示相应的方法：

(base) MacBook-Pro:~ ***$ brew -v
Homebrew 3.6.16-8-g895d322
fatal: detected dubious ownership in repository at '/usr/local/Homebrew/Library/Taps/homebrew/homebrew-core'
To add an exception for this directory, call:

	git config --global --add safe.directory /usr/local/Homebrew/Library/Taps/homebrew/homebrew-core
Homebrew/homebrew-core (no Git repository)
fatal: detected dubious ownership in repository at '/usr/local/Homebrew/Library/Taps/homebrew/homebrew-cask'
To add an exception for this directory, call:

	git config --global --add safe.directory /usr/local/Homebrew/Library/Taps/homebrew/homebrew-cask
Homebrew/homebrew-cask (no Git repository)

将上面提示的两段代码输入终端运行即可：

git config --global --add safe.directory /usr/local/Homebrew/Library/Taps/homebrew/homebrew-core
Homebrew/homebrew-core 
git config --global --add safe.directory /usr/local/Homebrew/Library/Taps/homebrew/homebrew-cask
Homebrew/homebrew-cask 

实战

我们先使用官网案例验证下：

import os
import tempfile as tmp
import warnings

from autoPyTorch.datasets.resampling_strategy import CrossValTypes

os.environ['JOBLIB_TEMP_FOLDER'] = tmp.gettempdir()
os.environ['OMP_NUM_THREADS'] = '1'
os.environ['OPENBLAS_NUM_THREADS'] = '1'
os.environ['MKL_NUM_THREADS'] = '1'

warnings.simplefilter(action='ignore', category=UserWarning)
warnings.simplefilter(action='ignore', category=FutureWarning)

import sklearn.datasets
import sklearn.model_selection

from autoPyTorch.api.tabular_classification import TabularClassificationTask

X, y = sklearn.datasets.fetch_openml(data_id=40981, return_X_y=True, as_frame=True)
X_train, X_test, y_train, y_test = sklearn.model_selection.train_test_split(
    X,
    y,
    random_state=1,
)

api = TabularClassificationTask(
    # To maintain logs of the run, you can uncomment the
    # Following lines
    # temporary_directory='./tmp/autoPyTorch_example_tmp_01',
    # output_directory='./tmp/autoPyTorch_example_out_01',
    # delete_tmp_folder_after_terminate=False,
    # delete_output_folder_after_terminate=False,
    seed=42,
)

api.search(
    X_train=X_train,
    y_train=y_train,
    X_test=X_test.copy(),
    y_test=y_test.copy(),
    dataset_name='Australian',
    optimize_metric='accuracy',
    total_walltime_limit=300,
    func_eval_time_limit_secs=50,
    #n_jobs= -1,
    memory_limit=None
)

api.search时，如果只加入了官网代码会报错

ValueError: current limit exceeds maximum limit

所以需要加上：

memory_limit=None

y_pred = api.predict(X_test)
score = api.score(y_pred, y_test)
print(score)

# Print statistics from search
print(api.sprint_statistics())

{‘accuracy’: 0.8728323699421965}
autoPyTorch results:
Dataset name: Australian
Optimisation Metric: accuracy
Best validation score: 0.8713450292397661
Number of target algorithm runs: 52
Number of successful target algorithm runs: 6
Number of crashed target algorithm runs: 45
Number of target algorithms that exceeded the time limit: 1
Number of target algorithms that exceeded the memory limit: 0

这样就运行完成了
下一节，讲一下具体的用法