scikit-learn: init your estimator lightly and fit heavily
If you run cross_validate(n_jobs > 1) with your own estimator, pay attention that your estimator will be copied per job.
最新文章
scipy: pdist indexing
scipy.spatial.distance.pdist(X) gives the pair-wise distances of X, $\operatorname{dist}(X[i], X[i])$ not included. The distances are stored in a dense matri...
scikit-learn: A walk through of GroupKFold.split()
Suppose $X[\text{“groups”}] = \begin{bmatrix} a \newline b \newline b \newline c \newline c \newline c \end{bmatrix}$ and n_splits=3.
Pandas: split a DataFrame into chunks
Prepare a test dataset:
Check the header of Columbia Eigen tab delimited files
Precomputed Eigen scores are available at Download Eigen/Eigen-PC scores.
R: Graphics Tricks
总结自 R Graphics Cookbook
SQL: Autocommit does not guarantee immediate committing
If you just google, you may find autocommit a vaguely defined term.
Python: pprint.pformat can be a good friend of str
```python import pprint
Python: Use datatime instead of time for local clocks
Effective Python Item 45: Use datetime Instead of time for Local Clocks