How to resolve numpy deprecation issue: AttributeError: module 'numpy' has no attribute 'str'. Did you mean: 'std'?

最近準備 Model Training Code 的時候, 發現幾個禮拜前的 Code 居然 Failed. 才發現一直沒仔細看 Warninig message.

FutureWarning: In the future np.str will be defined as the corresponding NumPy scalar. (This may have returned Python scalars in past versions.

1. 之前的版本

   $ pip show numpy
   Name: numpy
   Version: 1.23.5
   Summary: NumPy is the fundamental package for array computing with Python.
   Home-page: https://www.numpy.org
   Author: Travis E. Oliphant et al.
   Author-email:
   License: BSD
   Location: /home/jimmyliao/venvs/mlops/lib/python3.10/site-packages
   Requires:
   Required-by: contourpy, matplotlib, mlflow, numba, pandas, pyarrow, scikit-learn, scipy, shap
   

   原來, 一直都有提示即將被 deprecated.

   $ python train_iris.py    
   train_iris.py:28: DeprecationWarning: `np.str` is a deprecated alias for the builtin `str`. To silence this warning, use `str` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.str_` here.
   Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
     run.log('Kernel type', np.str(args.kernel))
   Attempted to log scalar metric Kernel type:
   linear
   /home/jimmyliao/workspace/lab/train_iris.py:29: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
   Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
     run.log('Penalty', np.float(args.penalty))
   Attempted to log scalar metric Penalty:
   1.0
   

2. 升級到目前的 release: 1.24.1

   $ pip install -U numpy
   Successfully installed numpy-1.24.1
   (mlops) jimmyliao@JimmyLiaoSL:~/workspace/lab$ pip show numpy
   Name: numpy
   Version: 1.24.1
   Summary: Fundamental package for array computing in Python
   Home-page: https://www.numpy.org
   Author: Travis E. Oliphant et al.
   Author-email:
   License: BSD-3-Clause
   Location: /home/jimmyliao/venvs/mlops/lib/python3.10/site-packages
   Requires:
   Required-by: contourpy, matplotlib, mlflow, numba, pandas, pyarrow, scikit-learn, scipy, shap
   

   執行就直接噴 Error

   train_iris.py:28: FutureWarning: In the future `np.str` will be defined as the corresponding NumPy scalar.  (This may have returned Python scalars in past versions.
     run.log('Kernel type', np.str(args.kernel))
   Traceback (most recent call last):
     File "/home/jimmyliao/workspace/lab/train_iris.py", line 60, in <module>
       main()
     File "/home/jimmyliao/workspace/lab/train_iris.py", line 28, in main
       run.log('Kernel type', np.str(args.kernel))
     File "/home/jimmyliao/venvs/mlops/lib/python3.10/site-packages/numpy/__init__.py", line 284, in __getattr__
       raise AttributeError("module {!r} has no attribute "
   AttributeError: module 'numpy' has no attribute 'str'. Did you mean: 'std'?    
   

3. 回頭找一下 Numpy 官網, 早就在 1.20 的文件就有說明.

   https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations

   Using the aliases of builtin types like np.int is deprecated

   For a long time, np.int has been an alias of the builtin int. This is repeatedly a cause of confusion for newcomers, and existed mainly for historic reasons.

   白話的意思就是 np.int 通常被拿來當作內建 int 的 alias, 在之前的版本其實都有 FutureWarning 只是都被忽略.現在要正式 deprecated.

   解決方法就照上面的文件說明, 取代成對應的. 像是:

   run.log('Kernel type', np.str_(args.kernel))
   run.log('Penalty', np.double(args.penalty))
   

   $ python train_iris.py
   Attempted to log scalar metric Kernel type:
   linear
   Attempted to log scalar metric Penalty:
   1.0
   Accuracy of SVM classifier on test set: 0.97
   Attempted to log scalar metric Accuracy:
   0.9736842105263158
   [[13  0  0]
   [ 0 15  1]
   [ 0  0  9]]    
   

收工.